Rick Anderson, University of Utah
Over the past ten years, the open access (OA) movement has achieved significant progress and high profile successes, especially in the United Kingdom and Europe, where centralised (and predominantly public) systems of higher education have made possible the establishment of mandatory OA policies at the national level. In the United States, where private higher education is more prevalent and where public institutions are governed at the state level, such mandates as exist are generally imposed by research funding bodies such as the National Institutes of Health, and usually require 'public access' rather than OA as strictly defined. (While some American colleges and universities have adopted institutional OA policies of various kinds, few if any of these can meaningfully be characterised as 'mandates'.)
The fact that so much of OA's growth has come as the result of mandates often passes unmentioned, but it is a fact that has real and sobering implications for OA's future. While there are certainly many individual scholars and researchers who have enthusiastically adopted OA practices on their own, their numbers are still dwarfed by those who have not done so, either because they have actively chosen not to, or simply due to some combination of inertia, disinterest, and unawareness of the options available.
That this remains the case some 20 years after OA first became a household word in scholarly communication circles is not particularly surprising or alarming – change is a slow process in our world – but it does raise two difficult, and possibly uncomfortable, but nevertheless very important questions for the future of the movement.
First: What percentage of scholarly and scientific authors will, if left to their own devices, ultimately choose to adopt OA?
Second: To what degree should that choice be left in the hands of authors?
The first question is, of course, unanswerable except by extrapolation and guesswork. The second is more interesting, and how it is answered by institutional decision makers, by funding bodies, by governments, and by authors themselves will have a heavy bearing on the future of OA. It is, of course, being answered piecemeal every day – every day that an OA policy of some kind is (or is not) adopted by an educational institution, funding agency, or governing body.
Before examining the question of authors' attitudes and rights in regard to OA, it is important to point out that this question is a separate one from that of whether or not OA is a good and desirable thing. For the purposes of this discussion, let us stipulate that OA is wonderful, that the benefits it provides solidly outweigh the costs, and that all scholars and researchers should adopt it. Unfortunately, this still leaves unresolved the question of whether those who produce scholarship should be compelled to adopt OA—and, if they should, by what mechanisms such compulsion can acceptably be applied.
In the course of my conversations with authors on these and related questions, I have noticed what seems to be a pattern. As authors express their reservations about OA mandates and about OA itself, four general categories of concern emerge.
Short for 'Creative Commons: Attribution', the CC BY licence is a public licence that allows anyone to "distribute, remix, tweak, and build upon your work, even commercially, as long as they credit you for the original creation". A great many authors oppose it, in part because it does not allow them to restrict commercial reuse of their work, and by bundling it into OA (either de jure or de facto, as in the case of the Berlin Declaration on Open Access), OA's supporters have put themselves into a very difficult situation. If 'real' OA means "OA with CC BY or its functional equivalent", then the only way to get most authors to go along is either to hide the fact that they're agreeing to CC BY (the fact that they are doing so is very often a surprise to authors), or by coercion. This may be an area in which OA advocates would be wise to accept compromise rather than pursuing a very strict orthodoxy.
Green OA competes with subscriptions
Scholarly authors are very often members of scientific and learned societies, many of which rely (to a greater or lesser degree) on revenue from journal subscriptions to underwrite services that their members value. Green OA – the model under which formally published articles from subscription journals are placed in public repositories – has not yet posed a significant challenge to subscriptions, largely because Green OA has not yet achieved pervasive success; relatively few articles can easily be found in their versions of record, in a timely way, in OA repositories. The more successful it becomes (meaning, the more high quality content is easily discoverable and freely available immediately upon publication), the more difficult it will become for libraries and other subscribers to continue justifying the cost of subscriptions. It is upsetting to some when this is spoken out loud, but it is an unavoidable reality, it is deeply concerning to many scholarly authors, and, at some point, it will have to be dealt with.
Author pays Gold OA is just another form of toll access
Subscription fees put a toll gate between the publisher and the reader; author fees put the toll gate between publisher and the author. (And while it is true that most Gold OA journals – or at least most of those listed in the Directory of Open Access Journals – do not impose author fees, most of those journals publish relatively few articles; thanks in large part to the massive output of the PLoS journals, which impose author fees and which together published roughly 35,000 articles in 2013, most Gold OA articles are, in fact, paid for by author fees.) For authors in the STEM disciplines, this means that research grant funds are redirected from research support to dissemination; for those in the humanities and social sciences, who are much more rarely supported by grant funds, paying fees is often simply not possible.
While the above concerns are mostly practical, this one borders on the religious: academic freedom is a sacred concept in higher education, and most scholars believe that it includes the freedom to decide whether, how, and where to publish one's work, and certainly the right to restrict commercial reuse of it. This may prove to be the most intractable barrier to comprehensive acceptance and implementation of OA initiatives; as long as we preserve the traditional understanding of academic freedom, authors will retain significant control over the distribution and licensing of their work. And to the degree that OA requires giving up that control, authors are very likely to resist it in significant numbers.
Faced with resistance on the part of recalcitrant authors, OA advocates have three choices: cajole, coerce, or compromise. For those invested in the goal of maximal OA, the first two strategies are risky: cajoling (or 'educating') is never 100% successful; on the other hand, coercion is fraught with both philosophical and practical difficulty.
That leaves compromise, a concept to which some some OA advocates are more open than others. It seems likely, however, that they will ultimately have little choice. As long as any choice is left to authors, a truly comprehensive OA future is difficult to imagine.