Can Open Access be done right?

Shortly before I wrote my last post about open access, I was given a copy of a very recent British Academy publication about open-access journals, and you may even remember that I cited it there.¹ I had, however, only looked at it briefly then and planned at that stage to write a sequel post using it to look at ways in which open access, which you will hopefully remember I don’t think has yet been developed as a working idea, might be. This is that post, but I can’t promise much by way of optimism…

The front cover

The book had an explicit brief from the British Academy, which was to evaluate how far any UK government or quasi-NGO policy on open access as a requirement for funding needed to vary across disciplines and what effect it would have on the UK academy to impose it (or, in the case of Research Councils UK, continue imposing the current one). All of this was more or less intended to settle some of the questions raised by a previous British Academy volume, and this one was explicitly focused on the situation in the UK. Though occasionally it looks across the Atlantic to the place where the results of the Research Assessment Exercise 2008 told the authors UK academics mostly publish when they don’t in the UK, and indeed compares [edit: the publication system] to the old Soviet Union on one occasion (note the third author), the conclusions and the dataset it presents on which those conclusions [edit: rest] only really apply in the country where I write.² There is an issue there which I’ll come on to but it’s an understandable restriction, and maybe it shows the way evaluations could go elsewhere.³

The other limit of the debate is that one of the main questions is taken as already settled out of court, that being the question of what type of open access we are debating. The last time I wrote about this I was cross about what has come to be called ‘gold’ open access, in which the publisher compensates for their loss of a product to sell by charging the author to publish with them, a charge (APC, article processing charge) that is usually thought will be supplied by the research’s supporting funding. At that point various voices were saying that for humanities research, often done without grants and equally often with very small ones, this was pernicious and would hit poorer institutions and younger students disproportionately. This is a position that the British Academy apparently took to be obvious and of which Research Councils UK has since come to be persuaded, and the result is that that is accepted as a model that only works for the sciences and perhaps only medicine (a position that the figures presented here justify) and that what we are actually studying here is ‘green’ open access, and exactly how to implement it.⁴ Obviously elsewhere that debate is not so finished, but this again may be something that this work could transmit to such fora.

The way that ‘green’ open access works, or is supposed to work, is that rather than charge the author, the publisher accepts that after a while it will put the work online for free, but it will not do this straight away, so that people who need the information as soon as possible will continue to buy the journal. They may also, when it finally goes online, only put the author’s submitted version online, which will not reflect subsequent changes or, obviously, correct page numbers, so it effectively can’t be cited. (Again, medicine has less of a problem with citing pre-prints, and I suspect that we will see more and more of this in the humanities, but for now it’s part of what gives journal publishers any hope and it has to be said (and is in this book, with figures) that basically almost no-one in the humanities actually puts up pre-print versions on the web anyway, Academia.edu or even personal web-pages not withstanding.⁵ Even I don’t, because how could you cite it? And so on.)

So with that accepted or assumed, the question becomes how long should the embargo period before the article is released to the world be? This is where the book is doing most of its work. In the first place, they show by an analysis of usage half-lives (a complex formula, given its own appendix, which tells you the median age of the content that made up half a journal’s downloads over a given period, and makes a reasonable index of comparison) that in general, the humanities do happily use content that’s older than medicine, but that actually, so does physics and most of the other sciences; medicine is just out by itself in its need to have the most immediate content straight away (and even there, the half-life figure was about six months on average).⁶ As they say several times, “the boundary does not lie between STEM (science, technology and medicine) and HSS (humanities and social sciences); rather, it lies between HSS plus Physical Sciences on one side and Medicine on the other”.⁷ The actual embargo periods being proposed as compulsory for humanities research funded by RCUK seem reasonable to them in the light of this, however, and so that ends there, and they go on to what is perhaps a more interesting set of questions about academic publishing more widely.

This is the point where I think there might actually be the sign of a set of answers emerging, at least for the time being, and it’s interesting. In the first place, they establish by means of a just-about-significant survey (Edit: 12% response rate! What can you do, though?) that librarians, who it is who actually buy journals, don’t pay any real attention to embargo periods when doing so and thus argue that publishers have nothing to fear from reducing them; and then they go on a two-chapter excursus about how journal publishing can and should be paid for, and this is one of my big questions about all such initiatives as you know so it made me read avidly.⁸ They don’t really have an answer, but what they show, by the same kind of back-of-the-envelope maths that I was using to disprove the possibility of crowdfunded higher education, is that it must be paid for, that only the smallest of journals can be run with no staff and no print costs and that as soon as one attracts any kind of following it needs an organisation that more or less amounts to a publisher. And since publishers need at the very least to pay for themselves, money has to come into the system somewhere, and whence is more or less an ethical debate depending on whom you think benefits most: the author, the academy or the world? And we might like to think it was the last, really, but the chances of any new tax revenue being put aside to fund open-access publication, as the authors here say, does seem fairly small.⁹ So we’re stuck in the middle with publishers and the only thing that matters, until that be solved, is how much libraries can afford to pay for journals and what publishers will charge for them. So I like this, obviously, because it more or less justifies my stance that even when the current academic labour of publication is uncosted, we can’t do this for free and have to answer the money question. What that means, in effect, is that whatever one’s ethical stance on open access may be, it is more or less irrelevant until we can come up with a better solution for academic publication than the current one, and that is a bigger problem than even three such sharp writers as these could be expected to solve in a 106-page volume, but it really needs solving.

I will permit myself just one of the various logos the open access movement has scattered across the Internet because I like the double signification of this one, it goes well with the post…

There are also some other important qualifications about coverage and inclusion here. Firstly and most obviously, this whole argument can only apply where publication is online. For the sciences that’s a no-brainer but looking over my own CV, of twenty-six outputs and seven reviews I could count over my career thus far, although six are virtual exhibitions and thus not only basically unimportant for research evaluations but self-evidently online, five of the reviews but only ten of the remaining twenty outputs are online automatically, seven of them behind paywalls, and three more are online because I put them there myself, not having signed any copyright away. My book is partly visible in Google Preview. The rest, ironically including quite a lot of the work about putting things on the Internet, is only available in hard copy, so remains very definitely closed. This is an issue the authors are aware of, substantially expressed as an awareness that electronic publication of actual books has a long way to go before it’s anywhere near general and that for most parts of the humanities, and especially the creative arts, that’s where most or much work goes.¹⁰ On the one hand this means that the figures and answers the authors come up with here are truer for psychology than any other HSS subject and affect, say, history, relatively little, but on the other hand means that if the less affected disciplines were suddenly required to make most or all of their research open access their publication plans would have to radically alter and would probably become partly impossible.

The other problem, and one to which the authors are alive in some ways, is that this really is an Anglophone and indeed UK problem. They emphasise that whatever the successes of the open access movement in the USA in creating impressive logos and impassioned stances (I editorialise somewhat), very few US publishers are paying any attention to it. They see this as a sign that what RCUK was proposing could seriously hurt UK academics’ ability to publish abroad.¹¹ I have tended to see it the other way, however, because of naturally looking at Europe. When I started my doctoral work basically no Catalan journal was online; now, almost all of them are, for free, open access. A goodly part of the French academic journal scene is also online via the Persée portal and there are German and Spanish equivalents too. Now it is certainly true that these are sometimes funded by the major state research organisations, because they publish most of the relevant journals; the fact still exists that the relevant state thought it worthwhile to fund that. In Catalonia, in fact, it isn’t even the state, but eighty-nine separate academic or learned institutions from museums and universities through to the Generalitat, which is funding it, but with the Generalitat one among many institutions contributing to it actually getting done. In these countries, someone did put aside tax revenue to present, organise and preserve academic research. Why we can’t, or won’t, do that, and why the justification of it is so much less obvious in the Anglophone world, not just to funders but to practitioners with our platitudinous explanations of the inherent worth of our subjects of study, is also quite an important research question, I’d say, even if not one I expect to see the British Academy funding however the results were published.

---

1. Rebecca Darley, Daniel Reynolds & Chris Wickham, Open Access Journals in Humanities and Social Science: a British Academy research project (London 2014), and it is of course, as you’d expect, online free and open-access, here.

2. The previous volume was Nigel Vincent & Chris Wickham (edd.), Debating Open Access (London 2013); comparison to the USSR Darley, Reynolds & Wickham, Open Access Journals, p. 85.

3. It should be remembered, though, that a great deal of the starter data here came from the UK’s Research Assessment Exercise 2008, without which the book probably couldn’t have been written, and certainly, without that or an equivalent, any country trying this will need to do much much more data collection. Of course, even that data was six years out of date by the time this book was published, and this is a fast-moving field, but since the Research Excellence Framework was only then being completed and has only just been counted, what could they do?

4. Darley, Reynolds & Wickham, Open Access Journals, pp. 16-20.

5. Ibid., pp. 71-74.

6. Ibid., pp. 49-66.

7. Ibid., pp. 8, 61 & 92.

8. Ibid., pp. 67-87.

9. Ibid., p. 84: “a frankly unlikely scenario”.

10. Ibid., pp. 24-32.

11. Ibid., pp. 33-35 & 36-48.