Copyright F.A.Q.
"What about the copyright issues associated with CogPrints? e.g. If I
want to publish a paper in a "normal" journal: do I have to ask the
journal's editor for permission to submit the paper (non-refereed and
refereed) to CogPrints?"
CogPrints is an author's archive; as such, it has the same relation to an author's work as the author's home institution does, when that work is archived on the home server (as all CogPrints authors are strongly advised to do, in addition to archiving it in CogPrints).

It is accordingly the author who must adopt a policy about copyright. We can only offer some generic advice:

    A distinction should be made between the unrefereed preprint and the refereed, edited, published reprint. No copyright agreement has any bearing on the unrefereed preprint, which can be publicly archived online before the refereeing even takes place.

    Hence the rest of the points below pertain only to the refereed, edited, published reprint. Preprints can be archived without reference to any copyright agreement or publisher.

      Note, however, that a minority of journals have indicated that they will not referee papers that have been publicly archived online. It is not clear whether any attempt has been made to enforce such a policy -- or indeed whether it would be possible to enforce it at all -- as so many authors are archiving their papers publicly on their home servers. See:

    If you have not signed a copyright transfer statement that cedes your right to publicly archive your own paper online for free, it is not clear that there is any problem, but if you wish to confirm this, you should inform your publisher that you wish to do so, and request confirmation that there is no legal obstacle. Some journals (such as all those published by the American Physical Society) explicitly permit public online archiving of the final published draft by the author; others attempt to specifically forbid it in their copyright agreements.

    Note that any copyright agreement pertains only to the final, refereed, edited draft that appeared or will appear in print. It does not, and cannot, cover pre-refereeing preprints or indeed any penultimate draft that preceded the final one. (The nature and size of the requisite difference between the two is to all intents and purposes arbitrary.) Nor does it cover substantive post-publication revised drafts, nor (most important) post-publication lists of "corrigenda" to the pre-publication preprint. This suggests a simple, legal way of getting around even the most restrictive copyright transfer agreement, as long as the author has already publicly archived the pre-refereeing preprint:

    If you have signed a copyright transfer agreement ceding your right to publicly archive your own paper online for free, you should contact your publisher indicating that you wish to do so; matters are evolving rapidly in this area and publishers may well be coming around to more justifiable and enforceable policies.

  1. You should always try not to sign restrictive copyright agreements. They are completely unjustified, they have nothing to do with protecting the author's intellectual property (for the refereed journal literature, which is and always has been an author give-away), they are only there to protect the publisher's revenue streams, and energetic steps are being taken to put an end to them as soon as possible. See the current copyright discussions and proposals in Science, Nature, American Scientist, and Chronicle of Higher Education, respectively:
a more detailed discussion about the copyright issues involved follows.

"Can I ask a journal to change their copyright policy?
Stefano Ghirlanda of Stockholms Universitet offers the following advice:
If you would like to ask a journal to modify their copyright policy so that you and others can post your articles on the web, you might find the following suggestions helpful.
Take the initiative

Many journals will accept a copyright agreement different from their standard one if asked to, but will not offer a liberal agreement from the beginning.

Thus, when you get the copyright-transfer form from a journal, just send back a different, already signed one with a science-friendly policy. You can model your requests after the American Physical Society's (APS) policy, which can be found at:

A possible sample text is:

I hereby transfer to [publisher or journal] all rights to sell or lease the text (on-paper and on-line) of [paper-title]. I retain only the right to distribute it for free for scholarly/scientific or educational purposes, in particular, the right to self-archive it publicly online on the Web.

More precise wording (legally speaking) can be found in the APS policy above. It should be made clear that only noncommercial distribution will be unrestricted, and that the publisher will retain all commercial rights.

In case of a "no"

If your agreement is declined by the journal, it may be helpful to express concern that an unjustifiably restrictive copyright policy hinders the free circulation of research findings. Point out that researchers' future willingness to submit to a particular journal could be influenced by its copyright policies.

Some journals are owned by Learned Societies, but the copyright is often managed by a commercial publisher. Try to go through the Society first, especially if you are or have been a member.

You can consider your time well spent even if the publisher fails to accept your conditions. It is important that the journals know what researchers are recognizing to be important preconditions for submission in the on-line era.

Stefano Ghirlanda Stockholms Universitet

Campaign for the Freedom of Distribution of Scientific Work:

Bachrach S. et al. (1998) Intellectual Property: Who Should Own Scientific Papers? Science 281 (5382): 1459-1460. September 4 1998.

"Any further advice?"
First I will tell you the correct answers, and then I will describe the potential complications, and how they are likely to be sorted out.
"Will inclusion of a paper in CogPrints count as 'publication' and hence render the paper no longer publishable in paper journals?
You must specify for/to whom it will "count"!

An Eprints Archive is certainly neither a journal nor a publisher; it is an on-line archive for providing free access to research, pre- and post-refereeing/publication. All authors retain the copyright for any papers they deposit and any unrefereed preprint that they archive in an Eprints Archive, they will  also (we hope), in parallel, go on to submit to a refereed journal, as they always did.

Appearing in an Eprints Archive is merely like appearing in a University Department's Tech Report Series (more and more of which are now electronically available too). For unrefereed preprints, CogPrints is just a global, collective Tech Report Archive for the Cognitive Sciences (but now globally integrable, through the Open Archives Initiative, into what is effectively a world-wide, pan-disciplinary, "virtual archive," free for one and all).

After that paper is refereed, revised and accepted by the peer-reviewed journal to which is has been submitted, the question arises about what to do with the archived preprint then: (1) Leave it in the archive, (2) withdraw it, or (3) supersede it with the final refereed, accepted postprint?

I assume that authors would overwhelmingly prefer (3), whereas journal publishers may well prefer it otherwise. Indeed, they may prefer it otherwise even for the preprint, not just for the published reprint... I will return to this.

The natural thing to do, and certainly the optimal thing for authors (if they overcome the restrictions of their journal publishers) when their article is accepted in this electronic age is to make the validated, peer-reviewed version as widely available as possible to everyone for free. There is no reason why its potential research impact should be constrained by fee-based access-barriers.

In the paper days we mailed reprints (for free); in the electronic age the natural thing to do is to email eprints. And there is no longer any need to go through the tedious reprint-requesting and reprint-sending phases; the eprint can simply be made universally available on the Net, either in the author's home institution's Eprint Archive (as part of the University's Electronic Reprint Series) or in a centralized, discipline-based Eprint Archive such as CogPrints (or both: Open Archive interoperability makes both solutions equivalent).

"Who retains copyright of the material? does inclusion in CogPrints imply implicit relinquishing of copyright by the author?"
The author retains the copyright. Nothing is relinquished, and authors will be expected to submit their papers to refereed journals in the usual way.

Where copyright issues will come up is on the question of whether authors should continue assigning full copyright to the publishers of the refereed journals, thereby in effect authorizing the publisher to restrict access to their papers (and hence restrict their potential research impact) only to those who pay to see them.

This is a huge conflict of interest, and it will be in this arena that the critical events will occur, not in the question of assigning copyright to electronic Archives: Unlike publishers, neither CogPrints nor University Eprints Archives has any financial interest in blocking access to the papers it archives, whether pre- or post-refereeing. Indeed, CogPrints has no financial interests at all, and Universities' interest, like the authors', is maximizing the potential impact of their research (the financial rewards come from the recognition and funding that results from the research's achieving its maximum impact)! The Eprint Archives are extremely cheap to maintain, and as their enormous potential contribution to learned inquiry and research impact is demonstrated, they will continue to be a part of Universities' infrastructure, just as the Web itself is.

I will come back to this when I discuss the APA's (American Psychological Association's) provisional policy on these issues.

"What kinds of papers would be included? whose papers would be included?"
An Eprints Archive is not, and is not intended to be, a refereed journal. Archiving of unrefereed preprints will be entirely under the author's control (although Archives will no doubt do some filtering to flush out the crazies). So anyone can archive anything, in principle, either in a centralized, discipline-based Archive such as CogPrints, or in their University's Eprints Archive. (All will be harvested into global "virtual archives" anyway.)

We have the precedent already of the Physics Archive, which has been public for over a decade and is already the locus classicus for nearly half the research in physics worldwide, and still growing. There are as many crazies in Physics as in any other discipline, but they have a negligible effect on the Archive.

Researchers searching the research literature will continue to seek the work of the authors they know and trust; moreover, there will always be two categories of papers in Eprints Archives: unrefereed preprints and, first, author-authenticated postprints of refereed, accepted papers (in the first phase it will be the author who "marks" them as such), and perhaps eventually also official, journal-authenticated postprints.

Now, to return to those questions that concern what existing journal publishers will do:

First, note that what the physicists have already done is again the harbinger of what will happen in all disciplines sooner or later. The inevitable and optimal outcome for researchers in all disciplines is that their refereed scientific and scholarly journals will abandon the trade model on which they are produced now. The trade model was unavoidable in the paper era, no matter how badly it served research and researchers: There was no choice but to treat refereed research reports (which are and always have been give-aways from their researcher/author's standpoint, not sources of sales revenue, which merely serves to limit their impact) in exactly the same way that books and magazine articles
(which are not, and never have been give-aways, but potential sources of royalty or fee revenue) were treated: The publisher charged readers for access to the "product," because that was the only way to recover the substantial cost of publishing in paper.

But whereas in trade publishing there is no conflict of interest between author and publisher -- both wish to receive payment for access to their product -- in refereed learned journals the author never sought (or received!) payment for his "product" (the report of his research). He merely went along, reluctantly, with the only viable economic model there was (namely, the trade model) if he wished to be published, and hence to have any research impact, at all.

The American Physical Society, the most prestigious publisher of refereed journals in physics, is working out with the Physics Archive a cooperative arrangement in which preprints can be submitted to APS for refereeing through the Physics Archive.  Once they have been accepted for publication by an APS journal, authors can self-archive their final, refereed draft, and if readers want to (pay to) see the "official" APS version (PDF), there is a link to that as well.

This partitions what used to be the classical publishing function into two separate components:
(1) the essentials, being the unrefereed preprint and the refereed postprint, which includes the all-important service of implementing peer review and (2) the optional add-ons, in the form of publisher's paper version, or the PDF (and whatever further value the publisher may add therein).

The devil is in the details, of course, because although the APS (like the APA (American Psychological Association), and the other APS -- the American Psychological Society) are learned society publishers rather than trade publishers, they must take steps to ensure that they are not ruined!

I believe those steps will be the following (but note that here I am trying to predict what the actual path to the inevitable and the optimal will be, so this is hypothetical, in contrast to the foregoing passages about freeing the refereed literature through self-archiving, which are not hypothetical):

Currently, the cost of publishing the average refereed article is about $2000. That is the total amount that the subset of institutions on the planet who are able and willing to pay the Subscription/Site-License/Pay-Per-View [S/L/P] costs for access to that article are currently paying.

The proportion of that cost that is dedicated to paying for the implementation of refereeing (because the peers in fact referee for free) is about 10% or $200 per (accepted) paper. If the publisher provides only that service -- let us call it Quality-Control and Certification [QC/C], certified by the journal's brand-name -- then that $200 per accepted paper can be paid for by the author-institution out of 10% of its annual 100% savings from the cancellation of all S/L/P costs. The refereed papers themselves will be free for everyone, online, in each institution's interoperable Eprint Archives, and the remaining 90% annual savings can be spent in other ways (e.g., on books, which are not, and never will be, author give-aways).

Or the 90% or even 100% can continue to be spent on the optional add-ons, via S/L/P -- but the refereed literature will all be free on-line either way.

That's what I (and an increasing number of others) think will be the inevitable and optimal solution. But what do publishers think, now? I append here, with comments, the APA's provisional policy on this. Let me note that they are not alone. The New England Journal of Medicine formulated a similar policy a few years ago, invoking, by way of justification, the "Ingelfinger Rule."

Let me add, though, that the APA, too, is us. It is up to us whether we comply with this provisional policy for the time being, for it is certainly a policy that is against our interests -- and one that could not be implemented if we chose not to comply.

I am not advocating a course of conflict with the APA. I am an advocate of quiet, sustained subversion rather than aggressive confrontation (for which there is no need, as the APS story is demonstrating). I think, therefore, that the best course of action is simply to ignore the following three "rules" that the APA has provisionally formulated (supposedly on our behalf, as we are they -- the editors, referees and authors -- and they are us):


American Psychological Association:

Posting Information on the Internet

So you want to put your article on the Internet on your own home page or
that of your university. What's the problem? There can be a few, especially
if you want to later publish the article or if it has already been

At their Spring 1996 meetings, the APA journal editors and the Publications
and Communications (P&C) Board took a long look at the implications of the
Internet for publishing.

* Notices in such journals as Neuroscience and the New England Journal of
Medicine alert authors that papers posted on the Internet are
considered already "published" and will not be considered for print

Now who is considering them as published, and will not consider them for print publication? Is it the readers of the APA journals? the authors? the referees? the editors?

No, for all four communities I have named are still perfectly capable of distinguishing unrefereed preprints from refereed articles, whether the distinction occurs in the on-paper or the on-line medium. The distinction is medium-independent and we all know exactly what it is.

Now in certain special cases, medical journals may have a real justification for warning their authors not to circulate unrefereed preprints, whether in paper or electronic form: Publicizing results that have not yet undergone peer review could be dangerous to public health.

But that concern is likewise medium-independent. Journals are free to set policies that are really intended to protect public health. But how much of the scientific -- or even the biomedical -- literature falls into that category? Cold fusion, for example, which was an instance of results that should not have passed peer review (although they did) was set aright much more quickly and effectively through the open unrefereed discussion and scrutiny it underwent on the Net!

Never mind: Let us concede that there are special areas in which it is justified to refuse publication if authors have self-publicized unrefereed findings that put public health at risk.

But what is at issue here is that vast preponderance of the research literature that does not fall in that category (and this includes the vast majority of the papers that appear in APA journals).

For that literature, the APA edict against archiving unrefereed preprints on the Net is unjustifiable and completely at odds with the psychological research community's interests: Would such a rule have been accepted if it forbade us to circulate our paper preprints [the ones we send around "for comments only, not for quotation or citation"], as we have been doing all along?

If the policy would have been seen as clearly unjustified (and unenforceable) in the paper medium, then exactly the same is true on the Net.

* Papers posted on the Internet may be considered in the "public domain"
and downloaded, incorporated into someone else's work and copyrighted
by them (i.e., authors can "lose" their own copyrights and their own
right to print publication).
I will not dwell on this statement. It is completely false, and by now I expect that there are many people in the APA who are embarrassed by it.

Statements like this have been made by the big commercial publishers, in an effort to intimidate their authors about posting their preprints on the Net. APA should be ashamed for doing likewise.

What underlies the effectiveness of this completely false statement is a confusion (in the minds of authors) between copyright as protection from:

(C1) theft of text, which, for the research community, is simply a victimless crime! every reprint-requester I send my paper to is not only "stealing" my text -- by not paying for it -- but I am actually paying [the offprint and postage costs] to aid and abet him in his crime! In short, this kind of theft is not what the nontrade author wants copyright protection from;
versus copyright as protection from:
(C2) theft of authorship, in which someone else publishes my work as his own. This IS something that authors very much want to be protected from (perhaps worrying more about it than is really necessary, but let's skip over that), but this is not at all what is at issue in archiving one's preprints on the Net!
The scare-tactic is to imply that posting on the Net is a threat to C2 (which authors certainly do not want) but in reality it is to prevent C1, which authors do want!

Nontrade authors must start to think more clearly about copyright.

* Posting a published paper on the Internet may violate the copyright transfer agreement related to the print publication.
It may indeed. And hence authors are well-advised not to sign such copyright transfer agreements. If there is a specific clause in a copyright agreement that implies that I cannot archive my paper on the Net, I strike that clause. If there is no such clause, I archive my articles on the Net just as I used to distribute reprints of my article.

But this is why I said that the copyright issue is not concerned with our archiving of our preprints on the Net -- we definitely retain copyright there -- but with assigning copyright to our publisher! Many people are now working out limited copyright arrangements with their publishers.

But there is no need for authors to wait for publishers to agree to more limited copyright arrangements. There is a completely legal way to self-archive one's refereed research even after having signed the most restrictive copyright transfer agreement -- as long as one has first self-archived the unrefereed preprint: Just append the "corrigenda": "The reader may wish to make the following changes in order to correct the preprint to make it conform to the refereed final draft"...

Many of the legal issues surrounding the Internet, online services, service providers, and copyright and privacy in the electronic world are murky, confused in the courts or soon to be there, and, of course, rapidly changing.
They are certainly murky and confused; and some parties are benefiting from keeping them that way: The nontrade author, who seeks and gets no royalties for his articles, has nothing to protect in court! He is not selling his texts, so he does not worry about, but rather applauds, their "theft"!

Copyright protection of authorship is, if anything, more enforceable on the Net than in paper. For although, to be sure, it is easier for me to pick up your text from the Net and pass it off as my own than it is to do it in paper, duplication of text is also much easier to detect in digital texts. So there's a trade-off: If I want to steal your paper text and publish it as my own, I have to scan it in with OCR, which is more labour-intensive than stealing bytes that have already been digitized, but it is probably also safer for me, because then I can publish it as my own in some other journal, and it is unlikely to be detected. Whereas on the Net, where gremlins are already scouring the entire world of digital texts as search engines 24 hours a day, my crime is much more likely to be found out.

Besides, this is nonsense anyway: Priority can be established much more rigorously and universally on the Net than in any other medium -- using encryption and date-coding.

So the legal issues referred to above concern only the trade  (i.e., non-give-away) literature, where it is the text that one wants to protect from theft; for us, nontrade (give-away) authors (or at least when we are wearing our refereed-research-reporter hats, for of course sometimes we do trade publishing
[e.g., textbooks, monographs, trade magazine articles] too!) it is only the authorship that we wish to protect from theft, and that is not at issue in any of this.

The APA P&C Board has, therefore, adopted an "interim" policy:

Authors are instructed not to put their manuscripts on the
Internet at any stage (draft, submitted for publication, in press,
or published). Authors should be aware that they run a risk of
having (a) their papers stolen, altered, or distributed without
their permission and, very importantly, (b) an editor regard such
papers as previously "published" and not eligible as a
submission a position taken by most APA journal editors.

This slightly garbled text is reproduced here verbatim. It calls to mind the sort of mumbling and stammering and coughing with which this sentence must have been uttered, because of course the editors, being us, have not taken this position, and certainly not on their own initiative. They have been drawn into it by the kinds of self-interested smoke-screen I have been trying here to dispel.

But I repeat that I prefer subversion to confrontation and conflict. I am sufficiently adept in this medium to be able to alert the vast preponderance of research psychologists, including, of course, those that are serving their terms as editors and associate editors (and referees, and authors and readers) of the APA journals. I could tell them all exactly what I have told you, and it would certainly make the scales fall from the eyes of many of them -- for very few of us have really given all this much thought: We are so accustomed to seeing the refereed journals as the disinterested arbiters of the quality of our research (as they certainly are) that we forget that there is also a trade side to learned journal publishing that has only an accidental relation to the invaluable service the journals perform for us -- and, in the on-line era, is actually in conflict with the best interests of research and research impact.

So if an author who knows that his job, his salary, his tenure, his likelihood of receiving research funds, prizes etc. all depend on the refereed journals, he will be easily persuaded that the "journals" are being as wise and disinterested when they formulate "interim policies" like the above one for us as they are in weighing the quality of our research.

But the source of this interim policy is not, and could not be, us (contrary to the case of peer review itself, where it is us). We are not saying that we regard preprints archived on the Web as publications, ineligible for refereeing; we are not the ones who say that upon acceptance for publication, we must not make our refereed findings public in the most powerful and ubiquitous medium that our species has ever created!

So you see that if I chose the polemical path, I would not be at a loss for words that would gird the research community into action.

But I prefer to take the peaceful path that the sensible physicists have quietly but firmly taken: to just go ahead and do it, and then let nature take its course.

Moreover, I hope that although what I've called "the inevitable and the optimal" solution for the learned research community will necessarily entail some downsizing by the big learned serials publishers -- especially the commercial ones -- I am still not among the most radical reformers in this area, who advocate that the peers should simply bolt from the existing journals and and re-establish their editorial boards, referees, and authors with new electronic-only publishers who are not already committed to the trade model that is at the heart of the profound conflict of interest that separates us from learned serials publishers at this critical juncture in the history of learned inquiry.

In addition, after acceptance for publication, the publisher is the copyright holder. APA does not permit authors to post the full text of their APA-published papers on the Internet at this time, as developments in the online world cannot be predicted. The APA will, however, closely follow such Internet developments. The P&C Board will establish a task force in June 1997 to investigate developments and recommend a longer term APA policy.
No doubt. But while there will no doubt be many developments for the task force to worry about, there is no reason for us to accept this arbitrary and unjustifiable stricture while the task force sorts things out! On the contrary, we should move vigorously toward the optimal and the inevitable, rather than allowing it to be needlessly delayed by those whose interests are so clearly in conflict with our own.

I recommend this because I really believe that it is our proceeding quietly but firmly along the path of subversion that will eventually bring publishers to their senses, and persuade them to restructure themselves in such a way as to continue to provide a service that is consonant with our interests, rather than contrary to them. For if an explicit confrontation is forced, I am afraid the traditional publishers are bound to be the losers.

"Thanks for the lengthy treatise on these matters. However, I still do
not feel that you have addressed the following question: Do you really
think that editors of virtual journals are likely to be able to
maintain the high standards of quality and moderate standards of
succinctness that characterizes our journals, when there are no
cost-enforced controls on the number and length articles in each
Actually, I think it will be the opposite: The very best journals will accept only the very best of articles, because not only will their page quota be unlimited in the positive direction (they will be able to accept as many articles as they like), but they will feel no pressure pushing them in the negative direction either: Paper journals need to fill their full annual page quotas or they lose those pages, which are budgeted and paid for. So on the Net they will be able to afford to accept more or less on the basis of quality alone, with no consideration of space.

In a nut-shell: Journal quality is determined by their rejection rate. The higher your cut-off point, the higher your quality, the bell-curve being what it is. This is medium-independent, and will be as true for on-line journals without page limits as it was for on-paper journals with page limits.

Do you really think it is space constraints that decide which articles go into JEP rather than QJEP? It's peer review; the referees do the evaluation and the editors set the threshold.

The reason the quality/prestige hierarchy among journals with different levels of refereeing rigor exists is not because the best journals have page quotas: it is because they have higher refereeing thresholds. There is no reason that Nature or Science (with their impact factors of 27.1 and 21.9) should lower their standards in the electronic medium; by the same token, why would we expect BBS (15.7) Psych Review (5.1) JEP:G (3.5), or QJEP (1.4) to lower theirs? In what sense is devaluing their quality-currency advantageous for learned journals, whose gold standard is quality, not quantity? Freed from page constraints and costs, what incentive is there to accept more than the best? Indeed, the Net will provide much more sensitive and informative ways of measuring the impact of authors and journals, which will always be there as feedback for journals should they inadvertently drop their guard.

In short, I think that, medium-independently, it is the level of rigor of peer review rather than space pressures that safeguard the quality of learned journals.

But for CogPrints, these questions are not even at issue, because CogPrints is not to be a journal but a preprint and reprint archive...

Below is the American Physical Society's Copyright form. As you will see, it does not rule out public archiving of the unrefereed preprint or the refereed reprint.

Note: This form is made available in electronic form for the convenience of authors, and the text should not be altered.

copy trnsfr.asc revised 6/00

Manuscript Number


The following transfer agreement must be signed and returned to the APS Editorial Office, 1 Research Road, Box 9000, Ridge, NY 11961-9000 before the manuscript can be published. Send requests for further information to the Editorial Director at the above address.

Article Title:

Names of All Authors:


Copyright to the above-listed unpublished and original article and subsequent, if necessary, errata, and the abstract forming part thereof, submitted by the above author(s) (the "Article") is hereby transferred to the American Physical Society (APS) for the full term thereof throughout the world, subject to the following rights that the author(s) may freely exercise and to acceptance of the Article for publication in a journal of APS. APS shall have the right to register copyright to the Article in its name as claimant, whether separately or as part of the journal issue or other medium in which the Article is included.

The author(s) shall have the following rights: The author(s) agree that all copies of the Article made under any of these following rights shall include notice of the APS copyright.

(1) All proprietary rights other than copyright, such as patent rights.

(2) The nonexclusive right, after publication by APS, to give permission to third parties to republish the Article or a translation thereof, or excerpts therefrom, without obtaining permission from APS, provided the APS-formatted version is not used for this purpose and provided the Article is not to be published in another journal. If the APS version is used, permission from APS must be obtained.

(3) The right, after publication by APS, to use all or part of the Article without revision or modification, including the APS-formatted version, in personal compilations or other publications of the author's own works, including the author's personal web home page, and to make copies of all or part of the Article for the author's use for lecture or classroom purposes. (4) The right to post and update the Article on e-print servers as long as files prepared and/or formatted by APS or its vendors are not used for that purpose. Any such posting made or updated after acceptance of the Article for publication shall include a link to the online abstract in the APS journal or to the entry page of the journal.

(5) If the Article has been prepared by an employee within the scope of his or her employment, the employer shall have the right to make copies of the Article for the employer's own internal use. If the Article was prepared under a U.S. Government contract, the government shall have the rights under the copyright to the extent required by the contract.

By signing this Agreement, the author warrants that this manuscript is original with the author and does not infringe any copyright or violate any other right of any third parties, and that the Article has not been published elsewhere, and is not being considered for publication elsewhere in any form, except as provided herein. If each author's signature does not appear below, the signing author(s) represent that they sign this Agreement as authorized agents for and on behalf of all the authors, and that this Agreement and authorization is made on behalf of all the authors. The signing author(s) (or, in the case of a work made for hire, the signing employer) also warrant that they have the full power to enter into this Agreement and to make the grants contained herein.

Author's Signature Date

Name (print):

If the manuscript has been prepared as a Work Made For Hire, the transfer should be signed by both the employee (above) and the employer (below):


Authorized Signature Name (print) Title Date


A work prepared by a U.S. Government officer or employee* as part of his or her official duties is not eligible for U.S. copyright. If at least one of the authors is not in this category, that author should sign the transfer Agreement above. If all the authors are in this category, one of the authors should sign below, and indicate his or her affiliation.

Author's Signature Institution Date (e.g., Naval Research Laboratory, NIST)

* Employees of national laboratories, e.g., Brookhaven National Laboratory, are not U.S. Government employees.


This is to certify that all authors are or were officers or employees of the Australian, Canadian, or UK Governments, which reserve their own copyright under national law. The following nonexclusive rights are hereby given to APS:

(1) The right to use, print, and/or publish in any language, form, or medium the Article or any part thereof, provided that the name of the author(s) and his/her/their foreign government affiliation is clearly indicated.

(2) The right to grant the same rights to third parties to print and publish the Article, subject to the provisions in item 1, above.

(3) The right to collect royalty fees.

Author's Signature Institution Date

Harnad, S. (1998-2001) For Whom the Gate Tolls? Free the Online-Only Refereed Literature. American Scientist Forum.

Harnad, S. (1998g/2000) The invisible hand of peer review. Nature [online] (5 Nov. 1998) and in Exploit Interactive 5 (2000)
Longer version:

Harnad, S. (1999) Free at Last: The Future of Peer-Reviewed Journals. D-Lib Magazine 5(12) December 1999

Harnad, S. (2000) E-Knowledge: Freeing the Refereed Journal Corpus Online. Computer Law & Security Report 16(2) 78-87. [Rebuttal to Bloom Editorial in Science and Relman Editorial in New England Journal of Medicine]

Harnad, S., Varian, H. & Parks, R. (2000) Academic publishing in the online era: What Will Be For-Fee And What Will Be For-Free? Culture Machine 2 (Online Journal)

Harnad, S. & Carr, L. (2000) Integrating, Navigating and Analyzing Eprint Archives Through Open Citation Linking (the OpCit Project). Current Science 79(5) 629-638.

Harnad, S. (2000) Ingelfinger Over-Ruled: The Role of the Web in the Future of Refereed Medical Journal Publishing. Lancet (in press)