On Open Access, Self-Interest and Coercion

On Thu, 24 Jan 2008, James J. O'Donnell [JJO'D] wrote (on liblicense):

JJO'D: "...Whether to include [books] in OA "mandates" is Stevan Harnad's question, and since I regard such mandates with skepticism, that question doesn't concern me."

But the question of mandates does concern a bigger and bigger constituency, now that the Australian Research Council, 6 of 7 UK Research Councils, the European Research Council, the Canadian Institutes of Health Research, the US National Institutes of Health, and a growing number of universities have already mandated OA self-archiving -- and the vast sleeping giant of universities worldwide is just about to awaken and follow suit:

22 funder mandates,
12 institutional mandates,
3 departmental mandates,

plus

5 proposed funder mandates,
3 proposed multi-institutional mandates,
1 proposed institutional mandate

That's a total of

37 mandates already adopted and
9 more proposed so far
= 46

So this might be an opportune time to re-examine the basis of one's skepticism about OA mandates...

JJO'D: "I am struck by the assertion that "all authors would want OA for their articles" if certain conditions are met. That's an interesting hypothesis, but I would simply underscore that the number of authors who currently do want OA for their articles is low enough that Harnad and others recommend they be coerced to achieve the goal. That fundamental disjuncture is important to understand and is advanced by empirical work, not by thought experiments."

(1) "Coerced" is a rather shrill term! (Is every rule that is in the public interest -- smoking bans? seatbelt laws? breathalyzer tests? taxes? -- coercion? Is academia's "publish or perish" mandate "coercion"?)

(2) It is empirically incorrect to assume that the number of authors that do want OA for their articles is the same as the number who spontaneously self-archive or publish in an OA journal today:

(3) Considerable empirical work has been done on these questions: The surveys by Alma Swan and others have repeatedly shown that (a) many authors still don't know about OA, and (b) many of those who know about it agree that they would want it for their articles, but they fear (wrongly) that it might be illegal, prejudicial to their publishing in their journal of choice, or just plain (c) too complicated and time-consuming to do it.

(4) As a matter of empirical fact, (a) - (c) are all wrong.

(5) More important, the surveys have found that although most authors still do not self-archive, 95% report that they would self-archive if their institutions and/or funders mandated it -- and 81% of them report they would do so willingly.

(6) In other words, most authors regard Green OA self-archiving mandates not as coercion, but as facilitation, for doing what they would want to do, but otherwise daren't (or otherwise could not assign it the proper priority in their academic publish-or-perish obligations).

(7) By way of still further empirical confirmation, Arthur Sale's many studies have shown that institutional self-archiving policies are successful -- and institutional OA repositories successfully approach capture of 100% of institutional research output -- if and only if they are mandates.

(8) All of that is empirical; there is one thought-experiment, however, and that is the various speculations and counter-speculations about whether or not Green OA self-archiving mandates will "destroy peer-reviewed journal publication" (see APPENDIX below).

(9) I fully agree that the only way to settle that question too, is empirically -- and the mandates will do just that.

(10) All indications are that if and when mandated Green OA should ever make the journal subscription model unsustainable, the only thing that will happen is a natural transition to Gold OA publishing, with (a portion of) the institutional subscription savings simply redirected to paying the (reduced but nonzero) costs of Gold OA: implementing peer review.

---

APPENDIX:

Would all peer-reviewed journal article authors indeed want OA for their published articles if they knew copyright was no obstacle and knew that self-archiving time/effort was trivial?

As noted above, I think we already know the answer to that question, indirectly, from the multidisciplinary surveys that have already been conducted and published. But suppose we wanted a still more direct answer:

Suppose we were to ask authors -- only about peer-reviewed journal articles (not books, irrespective of whether books are peer-reviewed) -- the following question (which needs to be as detailed as it is, in order to short-circuit irrelevancies, enthymemes, and incorrect assumptions):

"IF there existed no legal or practical copyright obstacles to doing it, and IF doing it involved negligibly little time and effort on your part (< 5 minutes of keystrokes per paper over and above all the time it took to write it), THEN would you be FOR or AGAINST making your own published journal articles Open Access so that all their potential users could access them, rather than just those whose institutions could afford paid access to the journal in which your article happened to be published?"

I am willing to wager that the vast majority of authors in all disciplines would reply FOR (and that if we added a box "if AGAINST, please state WHY," the reasons given by the minority who were AGAINST would all, without exception, be either factually incorrect, logically incoherent, or simply irrelevant).

That, I think, is the only real issue (especially given that a huge wave of institutional and funder self-archiving mandates is now growing worldwide: See Peter Suber's forthcoming SPARC Open Access Newsletter on February 2, 2008).

Some critics of OA mandates still seem to be seeing the self-archiving and the self-archiving mandate question as somehow imperiling the publication of articles in the author's peer-reviewed journal of choice: But articles published in the author's chosen peer-reviewed journal were part of the conditional (IF/THEN) in the above conditional question.

Hence any remaining reluctance about self-archiving can only be based on speculations ("thought experiments") about the future of journal publishing; those speculations would go into the "WHY" box, and then each one (they are all well-known by now!) could easily be shown to be groundless, empirically and logically:

(a) Self-archiving would bypass peer review. (Incorrect: The premise of the question had been that you deposit your published, peer-reviewed journal articles.)

(b) Self-archiving would jeopardize my chances of publishing in some journals, because they don't allow it. (Incorrect: (i) The premise of the question had been that there were no copyright obstacles. (ii) 91% of journals already endorse immediate self-archiving (for either unrefereed preprint, refereed postprint [62%], or both); for the 38% of postprints, the access can be embargoed and the IR's semi-automatic "email eprint request" button will provide almost-immediate OA to all eprint requesters during the embargo.)

(c) Self-archiving would jeopardize the survival of peer-reviewed journal publishing. (Un-empirical speculation: (i) There is no evidence to date that self-archiving reduces journal revenue, even in the fields where it has reached 100% years ago. (ii) Counterspeculation: If/when self-archiving made the subscription-based cost-recovery model unsustainable, peer-reviewed journal publishing would simply switch to the Gold OA journal publishing model, with institutions paying for their authors' publication costs out of (part of) exactly the same money they are currently using to pay journal subscription costs.

Stevan Harnad
American Scientist Open Access Forum