Harnad, S. & Hemus, M. (1997) All Or None: No Stable Hybrid or Half-Way Solutions for Launching the Learned Periodical Literature into the PostGutenberg Galaxy. In Butterworth, I. (Ed.) The Impact of Electronic Publishing on the Academic Community. London: Portland Press. 18-27. (Copyright 1997 Portland Press)


We are going to try to show why the future of scholarly and scientific research publication will not continue along the path that it has followed until the present day. The path it has followed until now is exactly the same one followed by every other kind of publication: best-sellers, textbooks, popular magazines. The written word has been treated as a product that is sold to reader-consumers according to the usual constraints of supply and demand.

And this is exactly the way the written word should be treated -- and will continue to be treated in the era of electronic publication -- when it is the author's wish that it should be so treated. Whether the medium is paper or electronic, the authors of best-sellers, textbooks, and newspaper and magazine articles will quite rightly continue to wish to be paid for their products, and will continue to seek copyright protection against the theft of their intellectual property: their words. In seeking remuneration and protection from theft, these authors will be making common cause with their publishers. Let us call the literature in question -- the literature in which what is optimal for the author is also what is optimal for the publisher -- the "trade" literature, although the "trade" descriptor is a bit misleading, because a good deal of what is called the scholarly/scientific literature would also be trade literature under the criterion we have proposed. The criterion bears repeating: it is when author and publisher are both agreed that the sale of their words and their protection from theft is what both of them want.

A lot of scholarly/scientific literature clearly meets this criterion: Scholars and scientists sometimes endeavour and succeed to write a book that sells well and brings them some royalty revenue. In fact, the litmus test for whether or not a text counts as "trade" literature is whether the author seeks and/or receives revenue from its publication and sale. From the author's standpoint, "trade" is not an insulting descriptor: After all, Nobel prizes have been awarded to poets and novelists whose works of art have not been any the less worthy for having been sold rather than given away.

One can say the same of newspaper and magazine articles written by scholars and scientists when they are writing for the popular press rather than their learned journals. There are magazines that are hybrid in this regard. The American Association for the Advancement of Science's journal, SCIENCE, carries both trade articles written by their staff science journalists for a fee, and research reports written by scientists for free.

For free. That's the key to the distinction we are making: What is the difference between these two different kinds of text in one and the same journal? Why do the research scientists not insist on payment for their contributions? Why do they instead, ready to suffer insult on top of injury, shower the editorial offices of SCIENCE magazine with ten or twenty times as many articles as are accepted for publication, in the fervent hope that their paper will be among the lucky few to be accepted to appear for free? It is not overstating it to say that if the authors of these nontrade articles could BUY their way into the pages of Science, they would be prepared to pay sizeable sums to do so!

What is the explanation of this anomalous behaviour? What is it that these authors want so badly that, far from wishing to be paid for the publication and sale of their words, they would be ready to pay to have them published and sold?

We all know the answer: The reward system of scholarly and scientific publication is not based on revenue from the sale of the author's WORDS, but on the importance and impact of the author's WORK. There is a "prestige" hierarchy among scientific and scholarly journals, a hierarchy that corresponds (for the most part) to the importance and soundness of the work reported in them. This importance and soundness is assured by the quality-control system of scholarly and scientific research publication, the system called peer review: Scholars and scientists evaluate the soundness of one another's work, first by providing informal feedback before it is submitted for publication and then by formally refereeing it when invited to do so on the basis of their expertise by the editor of the journal to which it has been submitted.

The peer review process differs from discipline to discipline. In some disciplines, the mark of excellence is their rejection rate, which can be as high as 90% (and probably higher in a journal like SCIENCE); in other disciplines, it is the acceptance rate that is 90% or more -- and this need not mean that the journal is of lower quality. Sometimes it is the very prestige of the journal that keeps contributors from submitting anything but the very best work to it for refereeing.

In any case, the quality of journals and their contents is well-known in the scholarly/scientific community, and on that quality are based not only decisions about what to read and trust but also about whom to hire and promote. This may go some way toward explaining the anomaly of the unpaid nontrade author's readiness to pay rather than seek payment for the appearance of his words in print.

Is that the end of it then? Is the status quo in paper simply going to duplicate itself in the new medium? To answer this question some words have to be said about costs: Why do the publishers of learned serials charge the amounts they do for individual and institutional subscriptions? There are many complaints about the strain that journal subscriptions put on University library budgets. Some say the publishers are charging more than they need to.

There is reason to believe, however, that supply and demand are operating naturally and properly in this market: When librarians approach their scholars to ask whether their serial holdings can be reduced, the scholars almost always insist that having the journals at hand, even if to consult them only rarely, is essential for their research, and this is probably not an exaggeration.

So if authors and readers are well-served overall by the state of affairs in paper, why not just recreate the same state of affairs in the electronic media? Paper publishers have all been, quite understandably, contemplating a hybrid transition scenario roughly like the following one: There are certain respects in which the electronic medium provides "added value" over and above the value provided by print on paper: An entire serial available on scholars' desks 24 hours a day, the possibility of doing sophisticated full-text searches, the availability of multimedia enhancements -- all these make the Net look like an attractive extension of the existing market.

So the transition should go roughly like this: Make an electronic version of the journal available to paper subscribers for slightly more than a paper-only subscription, and an electronic-only subscription for slightly less than a paper-only subscription. Then simply let things take their natural course: If and when the electronic medium prevails over the paper one, things will continue roughly the same way they had in paper: Subscriptions will still be paid for access to the journal, authors will be well-served as they had been, in the academic reward structure, and readers will also be well-served -- indeed better served than when it had to be their legs rather than their fingers that did the walking.

There are some variants on this ubiquitous transition scenario: Some contemplate a transition from subscriptions to site-licenses. Indeed, some form of site-licensing arrangement would have to be made even if the subscription model were retained, because an institutional subscription to an electronic version of a journal would have to be made accessible to all users at the subscribing institution, just as paid subscriptions are. The other contemplated variant is "pay per view," where it is every retrieval of an article that is charged for.

What all three of these hybrid scenarios have in common is that they retain the trade model: The learned author's work is still being sold in the same way that the trade author's words are sold. Or, to put it the other way, free access to the nontrade author's work is being denied just as it is denied to the trade author's words, until the access toll has been duly paid.

We have already described the anomalous reward structure of the academic author: Appearing in a prestigious journal is, to put it crassly, money in the bank. But it was always true -- ask any learned author -- that the fact that access to their work was restricted to those who paid was not one they particularly welcomed or drew any benefit from. Two practices on the part of learned authors that are doubly anomalous from the trade point of view were cues that something didn't sit too easily with them even in the Gutenberg era: Many authors purchased or made reprints at the time their paper appeared in print, ready to pay for that, plus the postage for sending them to readers who requested them. Imagine that! PAYING so that those who requested the paper could get it for free!

The reason for this largesse is no mystery: The purpose of publication for a scholar or scientist is not just to have his work appear in a prestigious journal. It is also to have his work read, used, built upon. The average paper among the 6500 journals indexed by SCIENCE CITATION INDEX receives no citations (if self-citations are not counted). This estimate could be off by a citation or two. It doesn't matter. The dividend for sending the paper out to those who request the reprint might be something spectacular: an increase of 100% or more in citations if the zero was an underestimate; an infinite increase if the zero is correct! Think about it.

The second anomalous practice by learned authors (and this is admittedly not common, but it makes you think) is the payment of page charges to have their accepted articles appear in paper more quickly. This really IS adding insult to injury: The funds are not from the author's pockets, to be sure. They are covered by the same grant or institution that supported the research. But the priorities from the author's standpoint are clear: The sooner their work is published, and the more it is read, cited and built upon, the better.

Those hiring and promotions committees are not indifferent to citations either, using individual citation statistics and journal impact factors to help judge the importance of their employees' work. None of this has any counterpart in the trade literature. Although the comparison is not flattering and should not be taken too literally, the learned author is as well-served by the levy of a charge for access to his work as an advertiser would be if potential clients had to pay to read his adverts. Think about it

Nor is it just in order to enhance citation statistics that the learned author craves interested readers. After all, whether or not publications and citations translate into career advancement, we must give learned authors some credit for having chosen learned inquiry as a career rather than, say, junk bonds. Scholars and scientists make their contributions to knowledge because they want their work to have an impact -- not the crass impact of citation statistics, but impact in the sense of making a mark on the course of human thought and knowledge. Any unnecessary barrier to access to their work is directly at odds with this desire. And subscriptions, site licenses and pay-per-view all erect barriers. So we must ask: are they necessary?

In the paper era, the answer would have been an unequivocal Yes. If paper serial publishers had not been able to charge for publishing the learned author's work, then the work could not have been published. The costs per page for publishing and disseminating paper periodicals were substantial ones, and it was only fair that the publisher should seek to recover his costs and a fair profit in exchange for the essential service that he was providing.

So although the access barriers erected by the publisher were means that were at odds with the learned author's ends -- namely, that all learned readers should have access to his work in perpetuum -- those means were also the only ones available. So these authors became accustomed to making the "Faustian Bargain" of assigning copyright to their publishers to allow them to protect their work from a theft against which the authors would have preferred no protection at all!

It is important to note that the publishers were not the devil in this Faustian pact; they were victims of the tyranny of print on paper and its economics just as the author was. If in place of paper there had been a viable way to extend the oral tradition, in the form of a global public address system on which learned authors could make public their findings for anyone interested, then the providers of the megaphone service would have been happy to charge for access to the megaphone on the transmitter's end rather than the receiver's.

Why was this not possible in the case of print on paper? Because the page costs were far too high for the author to afford them. Those costs had to be distributed across the learned community in the form of subscription fees. So the substantive question is: What would be the true cost per page in the electronic medium? Several expenses must be reckoned into this "first page" cost:

(i) The journal would have to implement peer review. Referees referee for free, but editors are usually paid a fee for the academic time they transfer to the journal. This would be a flat rate factored into the annual page costs.

(ii) The editorial office must process submissions, and referee reports, and provide copy editing, mark-up and proofs. This too would have to be factored into the page costs.

(iii) An important question is: Who should pay for the refereeing of papers that are NOT accepted for publication? In a journal with a 90% rejection rate this would not be a negligible sum. The choice would be to factor it into the peer review costs of accepted papers or to charge authors a refereeing charge (one that might help discourage unrealistic submissions!).

These are all controversial issues, but the correct answer depends on the true costs involved. Everyone would presumably agree that if the page cost for electronic-only publication with everything else factored in would be of the order of thousands of dollars per page, then paying them at the author's end would not be feasible. On the other hand, if it were one order of magnitude lower -- say hundreds rather than thousands of dollars per page -- then the benefits of ubiquitous unimpeded access free for all would begin to look more attractive.

Note that it is not only the author who benefits from the publication of his learned article: His institution benefits -- as reflected, for example, in the distribution of government research funds in the UK's Research Assessment Exercise, based as they are, on refereed journal articles in the sciences and in many nonscientific fields too [FOOTNOTE 1].

FOOTNOTE 1: In disciplines where the scholarly monograph is the measure of research productivity and impact, considerations similar to the ones we have raised for refereed journals are pertinent: The decision about whether or not to publish a scholarly monograph understandably depends at least in part on its likelihood of at least breaking even in sales. University publishers especially have been very generous in publishing esoteric monographs that were likely to have only a "succes d'estime." (Such efforts have, ironically, often been subsidised from journal subscription revenues as well as from profits earned by more trade-worthy scholarly books.) In the PostGutenberg era, however, it is no longer true that a scholarly monograph that receives a strong thumbs-up from peer reviewers but is not expected to generate enough sales to cover the page costs of paper publication must molder unpublished in a desk-drawer -- or, almost as bad, published (in paper) by a Vanity Press at the author's expense, with the associated stigma about its likely quality: Worthy esoteric monographs now have a new lease on life, without any compromise on scholarly rigour: peer reviewed for a distinguished publisher, they can take to the electronic skies bearing the same imprimatur as other scholarly works that happen also to have a wide enough market value to afford to appear in paper. (Even certain kinds of semi-refereed literature such as conference proceedings now have an alternative medium to appear in if they are not likely to be able to pay their way for paper.) [END FOOTNOTE 1]

In the case of scientific research reports, the tax-payer benefits too, in that he has already supported the conduct of the research; so if the work was worth supporting, its findings are surely worth reporting. And one hopes that society still regards itself as the beneficiary of scholarly work too, rather than just of scientific research.

In any case, there is a second source of funds to cover the reduced page costs of electronic publication: The same source that paid for them in print: the beleaguered library budgets released at last from the Faustian grip of paper subscription costs. There are still further sources one might mention, such as learned societies, but if the page costs are of the order of $4000 for a 20 page article, and what is at stake is worldwide access for free for all, there is hardly a contest over which option would be optimal.

The outcome is not only optimal; it is also inevitable. The only question is: How soon? and what is holding it up?

We have given up trying to predict when it will happen: Old habits die hard and there is no second-guessing human nature. Moreover, paper publishers can probably prolong the status quo for as much as a decade or so. By then, however, a new generation of scholar/scientists will be coming of age, one that feels comfortable with a screen and is accustomed to letting its fingers do the walking. They will have gone through their education taking the Web and its resources for granted. And there will be competition for their eyeballs. Authors will have been archiving their preprints and then their reprints electronically, and the new generation will of course learn that this is the fastest, easiest and cheapest way of accessing them.

Will publishers be able to invoke copyright to prevent authors from providing free refereed Eprints to anyone who is interested? This is highly unlikely, and the de facto indemnified status of Paul Ginsparg's revolutionary Physics Eprint Archive in Los Alamos, which is already the locus of over half the current physics literature and growing fast, confirms this.

Copyright only worked when both author and publisher sought protection from two kinds of theft: (1) theft of text and (2) theft of text authorship (in other words, plagiarism). As we have tried to show here, the first kind of theft, jointly condemned by the trade author and publisher is, on the contrary, condoned and even encouraged by the learned author. The second kind of theft -- plagiarism -- still requires safeguards, but those safeguards in the electronic medium will be technological ones such as encryption, date authentication, hidden attachments, and text-comparator engines that trawl the net searching for look-alikes. The Net makes it easier to plagiarise, but it also makes it easier to detect plagiarism.

Moreover, even theft of authorship would be a victimless crime with the average no-citation article: If the article is not found to be useful by the learned community under the name of its original author, it is unlikely to enjoy a better fate under another name! The true victim of stolen authorship is the plagiarist's employer, if it inspires a reward.

For important articles, their very importance will guarantee a high profile when they appear, providing an intersubjective date-stamping in the minds of its readers that will be backed up by objective date-stamping with at least as much reliability as printing it on paper and disseminating it worldwide had done in the Gutenberg era.

Can anything be done to hasten the advent of the optimal and inevitable solution in learned research publication?

Yes there is, and in a growing number of fields in physics, the paper house of cards is already poised to come tumbling down. As he readily concedes, the first author, with his "subversive proposal," has been "but playing John the Baptist to Paul Ginsparg's Messiah" (and historians will no doubt confirm this, once the optimal-and-inevitable has come to pass).

In 1994, Harnad circulated on the Net a subversive proposal intended to accelerate the collapse of the paper cardhouse. From his own experience of editing a paper journal, BBS, for twenty years, and an electronic journal, Psycoloquy, for seven years, Harnad knew what was holding contributors back. First, they were afraid that they would not get proper credit for publishing in an electronic journal; and second, the electronic medium did not appear to have the assured permanence of the paper corpus.

These fears were unfounded in principle, but self-fulfilling in practice -- as long as authors continued to balk at electronic-only publication.

So Harnad proposed a way that nontrade authors could have their cake and eat it too: (This proposal was the antithesis of the hybrid solutions contemplated by publishers that we described earlier). Learned authors should continue, as before, to submit their work to the prestigious paper journal of their choice but at the same time, they should publicly archive the unrefereed preprint on the Net. Then, once the paper was refereed, revised and accepted, they could swap that final refereed, edited version in place of unrefereed preprint on the Net.

The rest could be left to nature. The learned community could vote with their fingers and eye-balls, choosing the version they preferred. If that does not bring down the paper cardhouse, nothing will.

What about copyright, you ask? Well, several paper publishers have tried to stave off the optimal and the inevitable by warning their authorships that:

(1) Papers publicly archived on the Net would not be refereed or accepted for publication.

(2) Papers accepted for publication could not be archived on the Net.

And, for good measure, they warned their authors that:

(3) Archiving their papers on the Net made them public-domain and no longer the author's intellectual property.

We do not have the time now, as Fermat said about his Last Theorem, to rebut each of the points, but we have something better than a rebuttal: an existence proof: Paul Ginsparg's Electronic Archive in Physics is contradicting all three of these warnings and there is nothing anyone can do to halt its momentum and roll back to the days of toll barriers between the Physics community and the reports of their work.

The only difference between Harnad's notional subversion and Ginsparg's actual subversion is that Ginsparg's is centralised in one discipline-based archive, but distributed in multiple mirror-archives worldwide, whereas Harnad's way is more anarchic, calling for author-home-page-based Eprint archives, likewise woven together by links, mirror-sites and indexing agents.

Subversion probably works best if it is encouraged at both ends: centrally and individually.

We have since joined forces with Paul Ginsparg and will shortly announce a centralised Eprint archive for the Cognitive Sciences. Paul's archive is funded by a US NSF Grant and our Cogprints archive is funded by a UK JISC Electronic Libraries Grant. CogPrints will attempt to build on Paul's existence proof, generalising it to the multidisciplinary field of Cognitive Sciences, including Psychology, Brain Science, Artificial Intelligence, Robotics, Behavioural Biology, Linguistics and even parts of Philosophy.

The start-up funds for both Archives are intended to start the ball rolling; once they get enough mass and momentum, their course towards the optimal and inevitable for learned periodical publication will be unstoppable.

There are still some problems to solve, and we hope these will be addressed in other sessions at this conference. Chief among these are the potentially chaotic interim phase, in which libraries cancel subscriptions and some paper publishers go belly-up. This can be prevented if a rational and fair transitional scenario can be designed in advance: The key to cooperation is that the conflict of interest between the author and the publisher MUST be resolved. There is room for anything except the Faustian Bargain for Scholarly Skywriting in the PostGutenberg Galaxy.


Physics Archive: http://xxx.lanl.gov

CogPrints Archive: http://cogprints.soton.ac.uk

Ginsparg, P. (1994) First Steps Towards Electronic Research Communication. Computers in Physics. (August, American Institute of Physics). 8(4): 390-396. http://xxx.lanl.gov/blurb/

Harnad, S., Steklis, H. D. & Lancaster, J. B. (eds.) (1976) Origins and Evolution of Language and Speech. Annals of the New York Academy of Sciences 280.

Harnad, S. (1979) Creative disagreement. The Sciences 19: 18 - 20.

Harnad, S. (ed.) (1982) Peer commentary on peer review: A case study in scientific quality control, New York: Cambridge University Press.

Harnad, S. (1984) Commentaries, opinions and the growth of scientific knowledge. American Psychologist 39: 1497 - 1498.

Harnad, S. (1985) Rational disagreement in peer review. Science, Technology and Human Values 10: 55 - 62.

Harnad, S. (1986) Policing the Paper Chase. (Review of S. Lock, A difficult balance: Peer review in biomedical publication.) Nature 322: 24 - 5.

Harnad, S. (1990) Scholarly Skywriting and the Prepublication Continuum of Scientific Inquiry. Psychological Science 1: 342 - 343 (reprinted in Current Contents 45: 9-13, November 11 1991). ftp://ftp.princeton.edu/pub/harnad/Harnad/HTML/harnad90.skywriting.html

Harnad, S. (1991) Post-Gutenberg Galaxy: The Fourth Revolution in the Means of Production of Knowledge. Public-Access Computer Systems Review 2 (1): 39 - 53 (also reprinted in PACS Annual Review Volume 2 1992; and in R. D. Mason (ed.) Computer Conferencing: The Last Word. Beach Holme Publishers, 1992; and in: M. Strangelove & D. Kovacs: Directory of Electronic Journals, Newsletters, and Academic Discussion Lists (A. Okerson, ed), 2nd edition. Washington, DC, Association of Research Libraries, Office of Scientific & Academic Publishing, 1992); and in Hungarian translation in REPLIKA 1994; and in Japanese in "Research and Development of Scholarly Information Dissemination Systems 1994-1995. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad91.postgutenberg.html

Harnad, S. (1992) Interactive Publication: Extending the American Physical Society's Discipline-Specific Model for Electronic Publishing. Serials Review, Special Issue on Economics Models for Electronic Publishing, pp. 58 - 61. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad92.interactivpub.html

Harnad, S. (1995a) Electronic Scholarly Publication: Quo Vadis? Serials Review 21(1) 78-80 (Reprinted in Managing Information 2(3) 31-33 1995) http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad95.quo.vadis.html

Harnad, S. (1995b) Implementing Peer Review on the Net: Scientific Quality Control in Scholarly Electronic Journals. In: Peek, R. & Newby, G. (Eds.) Electronic Publishing Confronts Academia: The Agenda for the Year 2000. Cambridge MA: MIT Press. Pp. 103-118. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad95.peer.review.html

Harnad, S. (1995c) The PostGutenberg Galaxy: How To Get There From Here. Information Society 11(4) 285-292. Also appeared in: Times Higher Education Supplement. Multimedia. P. vi. May 12 1995. http://cogsci.soton.ac.uk/~harnad/THES/thes.html

Harnad, S. (1995d) There's Plenty of Room in Cyberspace: Response to Fuller. Information Society 11(4) 305-324. Also appeared in: Times Higher Education Supplement. Multimedia. P. vi. June 9 1995. http://cogsci.soton.ac.uk/~harnad/THES/harful1.html

Harnad, S. (1995e) Universal FTP Archives for Esoteric Science and Scholarship: A Subversive Proposal. In: Ann Okerson & James O'Donnell (Eds.) Scholarly Journals at the Crossroads; A Subversive Proposal for Electronic Publishing. Washington, DC., Association of Research Libraries, June 1995. http://cogsci.soton.ac.uk/~harnad/subvert.html

Harnad, S. (1995f) Interactive Cognition: Exploring the Potential of Electronic Quote/Commenting. In: B. Gorayska & J.L. Mey (Eds.) Cognitive Technology: In Search of a Humane Interface. Elsevier. Pp. 397-414. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad95.interactive.cognition.html

Harnad, S. (1997) How to Fast-Forward Serials to the Inevitable and the Optimal for Scholars and Scientists. Serials Review. (In Press) http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad97.learned.serials.html

Harnad, S. (1996) The Origin of Words: A Psychophysical Hypothesis In Velichkovsky B & Rumbaugh, D. (Eds.) "Communicating Meaning: Evolution and Development of Language. NJ: Erlbaum: pp 27-44. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad95.word.origin.html

Harnad, S. (1997) The Paper House of Cards (And Why It Is Taking So Long To Collapse). Ariadne 8: 6-7. http://www.ukoln.ac.uk/ariadne/issue8/harnad/

Hayes, P., Harnad, S., Perlis, D. & Block, N. (1992) Virtual Symposium on Virtual Mind. Minds and Machines 2: 217-238. http://cogsci.soton.ac.uk/harnad/Papers/Harnad/harnad92.virtualmind.html

Odlyzko, A.M. (1995) Tragic loss or good riddance? The impending demise of traditional scholarly journals, International Journal of Human-Computer Studies (formerly International Journal of Man-Machine Studies), 42 (1995), 71-122. Condensed version in Notices of the Amercan Mathematical Society, 42 (Jan. 1995), 49-53. Available at URL http://www.research.att.com/~amo/doc/complete.html