Open Access publishing: A Personal View
This is a guest post written by Sara Delamont and Paul Atkinson, Editors of the journal Qualitative Research that sets out their response to the recent Finch Report and its implications. We welcome further blog posts from socialsciencespace members and the wider social science community on responses to this important issue.
As will be seen from our comments below, we are not convinced by many aspects of the current proposals in practice. We are also far from convinced by the general rationale for open-access publishing, especially the ‘gold’ option, under which the author (funder, institution or individual) pays. We recognise that the Report of the Working Group chaired by Janet Finch is by no means the only contribution to the current debate. We are also aware that SAGE in common with other major publishers has been contributing to the relevant discussions. We are personally sceptical about many of the major premises of Open Access Publishing, and we are certain that many of the potential implications have not been thought through. The Finch Report pays remarkably little heed to the detailed arrangements that may need to be put in place, or many of the potential consequences for individual scholars, research groups and academic departments.
The imperatives for Open Access publishing are poorly articulated, and lack logic. It is asserted that the results of publicly-funded research should be more widely accessible. In point of fact the ‘findings’ of research are almost always freely available, as investigators are required by funders and others to disseminate their results widely (public-engagement events, web-pages, press-releases, media exposure etc.) In fact, Research Councils and major charities, such as Wellcome, could overnight make the results of their funded research publicly available by publishing final reports of all funded research projects on their websites, after the external evaluation process is completed. This, of course, would not satisfy the less overtly articulated agenda of trying to clip the wings of commercial publishers. There seems to be a political campaign based on a desire to stop public money, in the form of university library subscriptions, going to commercial academic publishers, while some academics are apparently adopting a characteristic, would-be patrician disdain for ‘trade’. In the current political climate, however, it seems odd. Vast amounts of ‘public’ money are in fact channelled into the ‘private’ sector. Through procurement and out-sourcing large swathes of public services are in private, for-profit hands – from prison services to defence procurement, to pharmaceuticals and healthcare provision. There seems no special reason to disrupt the long-standing ecology of publishers and academics. Open Access publication will only guarantee universal access to research, of course, if all the papers that are submitted are actually accepted for publication. The only alternative will be to expand open-access journals until they accept 100% of all papers submitted to them, which would of course destroy their credibility. (But the financial temptation is clearly there if every paper comes with a £1,000 to £2,000 price-tag,)
There is little evidence that there is a widespread demand for the technical contents of the 25,000 academic journals, while the evidence from hits and downloads of electronic journal papers suggests that a very wide readership is already being reached. There is an analogy here with the ESRC data archive. Reviews showed that there were many data-sets deposited there that were never used for secondary analysis, while a small number of national data-sets accounted for virtually all of the usage. The current public discourse on open-access publishing is – as is so often the case – predicated on a model derived from the natural and biomedical sciences: externally funded, with publications that report the key findings or research programmes supported by long-term funding, conducted by large and relatively stable research groups etc. Large-scale STEM is research is concentrated in a relatively small number of well-found universities and departments. By contrast, social and cultural research is much more widely spread, much of it undertaken in institutions without QR funding, and without external research grants. In the social and cultural disciplines, publication is often more open-ended, takes place over a longer time-period, and is often much more episodic in nature. STEM subjects normally have ‘findings’ with a short shelf-life that are published within the life of a project, while the shelf-life of social data is much longer.
Issues for authors and journals
Much of the research in social and cultural disciplines is not based on large-scale, externally-funded research grants that can (with some problems) fund author-pays publishing. Much research and scholarship is done on an individual basis, with little or no direct funding. (Of course, it is nominally paid for from the 40% or so notionally available to a tenured academic on a teaching-and-research contract in an HEI unit of assessment that receives QR funding). Who pays for its publication? Clearly it has to be the employing institution – from departmental allocations. But who decides how many papers each scholar is entitled to publish on this basis? Again, is this an open-ended commitment? Clearly not, when budgets are finite. We know that some of the authors we publish do so in an environment that is hostile to qualitative research – especially in the United States, where tenure committees, deans and chairs of department can take a strong line. At the moment there is no bar to anyone submitting and publishing work on any topic, in any intellectual style, from any methodological or conceptual standpoint. We do not envisage that lasting. The consequence may not even be completely unintended.
In disciplines such as anthropology and sociology, the fruits of research (which may have extended over a period of time) are often contained in scholarly monographs. Whether published by university presses or commercial publishers, these are behind the pay-wall, being expensive (say £60) and mainly purchased by university libraries and a small number of committed specialists. Does such publication also fall under the same criticism? Will funded researchers be prohibited from publishing monographs? Must they show (to whom?) that the ‘findings’ (whatever they are) have first been made publicly available in open-access journals? (And if so, the monograph becomes redundant, even though it is the right way to convey many social and cultural studies).
When publication has to be paid for from institutional funds, who will decide how many papers any given academic can and should publish? Ten a year? Five a year? Two a year? For how many years? Perhaps each academic staff member will have to be given an allocation by their department, from a pot of money set aside for the purpose (presumably saved from the fund that used to pay for journal subscriptions). But will decision-makers (e.g. Research Committees, Deans or Heads of Department) then prioritise certain journals over others? On the basis of impact factors, perhaps, or on the basis of blind prejudice. Who knows? Or perhaps priorities will be set on the basis of the subject-matter of the papers and journals themselves. Will oral history be given the same priority as econometrics? Will curiosity-driven research get the same funding as pragmatic research with guaranteed ‘impact’? Colleagues in LGBT studies already report that senior academics are ‘hostile’ to their work, and other unfashionable or controversial topics could be censored. The current freedom to publish in whatever journal is deemed appropriate by the author will be threatened in some quarters. The funding of publications will be a very ready mechanism to extend further the current utilitarian approach to research while further marginalising work already under-valued. Will individual academics be able to pay for their publications themselves if nobody else will fund it? Will HMRC or the IRS allow publication fees as tax-deductible? Some individuals might be able to subsidise their own publications – though how many would be willing to do so is moot – but most would not; this is also potentially divisive. While Research Councils and other funders might commit themselves to funding the publications of their doctoral students, many are part-time, and are not funded by any funding body. Who will pay for their papers?
On the basis of informal conversations both in the UK and recently in the US it is abundantly clear that people are not fully aware of the current climate, and have certainly not thought things through from their own perspective. Our own view is that open access publishing where the author pays is likely to be highly divisive, will undermine yet further academic freedom in some quarters, and will have unintended consequences of suppressing some work rather than rendering it increasingly available. Our primary objective should be to ensure that research is published solely on the basis of quality, and not on other methods of selectivity.
Our own back-of-the-envelope estimates suggested to us that £1,000 per article was likely to be the lowest viable fee for publication. We strongly suspect that – for reasons given above – this will be a serious disincentive to many individuals who are not able to secure institutional funding. It is also sufficiently high for there to be real issues of rationing within and between academic departments.
Many journals are owned by learned societies, even when they are produced and managed by commercial publishers. They are often the only real material incentive that accompanies membership and they underpin the regular income stream that membership subscriptions provide for the society. If the contents are universally available free to readers, it is not clear what benefit resides in membership. There is a clear threat to the sustainability of learned societies here. We also note that at present there is only one material incentive for editorial board membership and the consequent obligation to referee papers – a free subscription to the journal. Open access publishing would remove that incentive. Publishers may need to consider what inducements might be offered instead: perhaps a fee-waiver for a set number of publications in one of their journals? Would there be any such material incentive for individuals not on the board to act as reviewers?
The devil is in the detail, and so we consider some possible scenarios in more detail. We continue with a specific case-study: Paul Atkinson has recently co-authored a paper on animal models in biomedical research, forthcoming in Sociology, a major BSA journal that is an important income stream for the professional association. The paper derives from funded research in Cardiff and in Exeter. There are four authors. There are two institutions. The authors are in four different academic departments. The paper derives from three different funded research projects. All of the separate projects were completed some time ago, and the specific financial accounts will have been closed. The paper is a conceptual and empirical analysis of a specific topic common to all three projects. It was never planned as a specific output of any one of the original research projects. While the research projects were all funded – by ESRC and by charities – it is far from clear who, under author-pays systems, would be willing and able to pay for this particular publication, and sorting out the financial details would seem like an administrative nightmare. Would the payment have to be agreed and committed before submission? Or after final acceptance of the paper? How many months would be added to the cycle of submission, revision and resubmission, editorial decision-making and final publication? Publication in the social sciences can already be a protracted process. We are aware of journals (not our own) where the initial editorial decision can take up to a year, and decisions on revised versions can add yet another. (It would be a greater service to research to speed-up publication times rather than disrupting the entire system.)
In the social and cultural disciplines, this is not an especially extreme case. There are many publications that are (a) unforeseen, and (b) retrospective, in the sense that they are drafted some time after the official end of a specific project. Paul Atkinson, for instance, published a paper in 1992, in Qualitative Health Research, based on a secondary analysis of data collected for his PhD which was submitted in 1976. Sara Delamont published a paper ‘Who needs the leotard? ‘in 1988, based on a re-analysis of data collected for the ESRC-funded ORACLE project which ran from 1975 to 1981. Those oracle data were also drawn on for a joint paper with David Mellor that was published in 2011, and that also used data from Mellor’s 2003 ESRC-funded PhD thesis. Sara Delamont’s current work on martial arts is not externally funded, and the fieldwork is in effect done in her own time. It has been published in a wide variety of top journals. Who is to say how many she is entitled to publish at her institution’s expense?
In other words: We have experience of large-scale funded research projects. We have directed ESRC and Wellcome Trust research. Such funded research seems in itself to be relatively unproblematic: the grant will include a budget-line for publication in open-access journals. This already happens. But even projects of that sort can throw up problems. How long after a piece of research is completed – in the sense that the final report has been accepted and publications of record have reported the main ‘findings’, is there an obligation or a commitment to pay for publication costs? If funders are willing to pay for the publication of ‘findings’, are they equally willing to pay for methodological or conceptual pieces that relate to projects, but are not confined to the empirical research itself? What are the priorities, and how are they to be set? Not every intellectual output can be predicted in planning and costing a single research project.
Immediate issues for SAGE and our own journal
We recognise the reality of the current situation, however, and therefore we now comment on the alternatives open to us and to SAGE. Clearly the option of offering free access via SAGE Journals Online is a sensible compromise, given that (at present) a few authors may be obliged to do so as a condition of their funding. Clearly this can be incorporated into current and future editorial policies, and provided it is activated post-acceptance, then the impact is likely to be negligible in the immediate future. The up-take of this option needs to be monitored closely, not least as part of the evidence-base used to evaluate the impact of open-access publishing.
We do not believe that the embargo period under a ‘green’, repository-based system is especially sensitive in disciplines like our own. It is clear from the available evidence that citations to journal papers cumulate over lengthy periods after publication. Papers in QR, for instance, do not contain findings that are time-sensitive. We believe, therefore, that the current arrangements are appropriate. We are compliant with green open-access conventions and we see no reason to depart from that.
We do believe strongly that the full implementation of any policy, especially during any ‘transitional’ period needs careful monitoring and the collection of evidence. It will be apparent from our comments above that we are not convinced by the current arguments. The number of hits and downloads globally for even a specialised journal like QR does not suggest that its contents are invisible to a large world-wide community of scholars who are disenfranchised by the current arrangements. Such baseline data need to be used to evaluate the impact of open-access publishing. We shall also need good evidence on patterns of submission (country of origin, status of authors, institutions of authors etc.) as well as global measures of volume, if we are no monitor the longer-term impact of change.
We have not so far commented on institutional support for journals themselves. Hosting the editorial office of a well-respected journal is something that university departments often choose to subsidise. After the 2001 RAE, for instance, our own department subsidised a major journal to the tune of £19K to £25K per annum, in order to gain from the prestige of hosting the journal ahead of the 2008 RAE. At minimum they provide office-space, cleaning, heating and light, and other facilities at no direct cost to the publisher. Commercial publishers rarely pay the full direct and indirect costs of journal editorship. If the entire funding basis of journal publishing is to be reviewed, then universities themselves may start to apply more rigorous costings themselves. We know from the implementation of full economic costing regimes for research and other services that this can increase the overall costs considerably (in some cases doubling them).
Finally, we note that publishers will still be for-profit companies, who will decide what is an appropriate level of profit? Who will therefore decide on the appropriate fee level per paper to be levied on universities? Will we now have to have yet another market regulator? And will it be called Offprint?
Read Related Articles