Re: Archiving Journals in CogPrints

From: Stevan Harnad <>
Date: Mon, 12 Aug 2002 00:11:37 +0100

On Sun, 11 Aug 2002, Manfredi M.A. La Manna wrote:

> The current (and fluid) situation is slightly more complicated than Stevan
> makes out. Even publishers who appear to take a liberal view on
> pre/postprints do impose rather tight restrictions when you read the small
> print. Take the following example:
> Publisher X:

First of all, why "Publisher X"? Is this a real example, or a
hypothetical one? If a real example, why conceal the publisher's name?
Is there any reason why any publisher's copyright policy should not be

If this is a hypothetical example, why are we bothering with it, instead
of considering real examples?

I shall continue as if this were a real example.

I note also that Manfredi is conflating copyright matters and
embargo/Ingelfinger matters willy-nilly in his postings. I will
try to tease them apart as we go along:

> "authors who assign their copyright to us retain unlimited
> free reproduction rights for their own work. Authors do not give up their
> rights to republish or reproduce their work for course notes, in another
> journal or as a book chapter, or electronically including their own or
> institutional Web Site, subject to acknowledging first publication details."

The "first publication details" referred to here are the details of the
publication in the journal itself! "When you re-publish (or self-archive)
the published article, please include the first publication details,
clearly indicating that it was published first in this journal."

And this passage is about the post-publication self-archiving
of the refereed postprint, hence a copyright matter, not about the
pre-submission self-archiving of the unrefereed preprint, which is merely
a journal-policy (Ingelfinger/Embargo) matter.

I would be quite happy if all 20,000 peer reviewed journals adopted the
above copyright agreement for the self-archiving of refereed, published

> This seems quite self-archiving friendly. However, authors have also to
> sign an undertaking that
> "we warrant that the Work has not been published before".

Of course, and it is most reasonable that a journal should request this.
Why should it waste its referees' time and its journal's space on papers
that have already been published in other journals (or books, or
conference proceedings)?

> In other words, to quote from the explanation given by the editor of a
> journal published by X:
> "The freedom to use your paper in whatever form kicks in *after* the
> publication date of the paper in the journal, not before.
> Pre-publication on a website would infringe this because a guarantee has
> been signed that "the Work has not been published before".

With all due respect to the anonymous author of this explanation of this
anonymous journal's copyright policy, this is nonsense!

As I have pointed out many, many times before, the legal definition of
"publication" is already fulfilled by committing a text to paper once and
showing it to someone. It is fulfilled a fortiori by circulating paper
preprints to colleagues for comments. Therefore this legal definition of
"publication" cannot possibly be the one on which copyright transfer
is based, otherwise the manuscript on which a paper is submitted to be
considered for publication would already constitute a prior publication,
and hence make ALL submissions ineligible!

Hence logic dictates that this abstract sense of prior "publication"
cannot be the one meant in the passage in the copyright transfer
agreement attesting to the paper's not having been previously published.
What is meant by prior publication here is formal publication in a
journal, book, or conference proceedings, i.e., under a publisher's

Having followed the logic so far, it should not require much effort to
see that "prior publication" cannot, by the same token, apply to the
prior informal circulation of eprints either, whether by email or via
the web (especially in these days of email and web submissions!).

I would suggest that this sort of foggy thinking is not worth our
attention, even if it comes from an anonymous editor "explaining" an
anonymous publisher's copyright policy. If you want legal advice about
the meaning of contracts, don't ask an editor! (This advice, however,
is a bit tainted by the Cretan Liar paradox...)

Here is what does make sense, and what is meant by such copyright
transfer agreements:

    "We will not re-publish your paper in our peer-reviewed journal if
    your paper has already been published before (in a peer-reviewed
    journal, or a book, or conference proceedings). Prior
    informal circulation of the unrefereed preprint (on-paper or
    on-line) does not count as 'publication'."

End of story insofar as copyright transfer contracts and preprints
are concerned.

The Ingelfinger Rule (see ).
goes further. It says:

    "We will not even REFEREE for potential publication a paper that
    has been made public (in any form -- whether an online preprint or
    even a press release) prior to its ACCEPTANCE for publication by
    this journal."

Preprint self-archiving would indeed be a "violation" of the Ingelfinger
Rule. My advice to all authors is to go ahead and violate the Ingelfinger
Rule! It is both indefensible and unenforceable, it is not even a
legal matter, and it is on the way out anyway:

    Harnad, S. (2000) Ingelfinger Over-Ruled: The Role
    of the Web in the Future of Refereed Medical Journal
    Publishing. Lancet Perspectives 256 (December Supplement): s16.

    Harnad, S. (2000) E-Knowledge: Freeing the Refereed Journal
    Corpus Online. Computer Law & Security Report 16(2) 78-87.

> Notice that if a form giving the above guarantee is not signed by the
> author, the paper is not published. Full stop. So much for
> "the Ingelfinger Rule [being] virtually unenforceable".

Once the paper whose preprint has already been self-archived has been
refereed and accepted for publication, the Ingelfinger Rule has already
been violated! Before any copyright transfer statement has even been

Manfredi is conflating the Ingelfinger Rule (not a copyright matter) with
the above-discussed misconstrual of what constutes "prior publication"
for purposes of copyright transfer (a copyright matter).

Sign the copyright transfer form (if you must), and don't worry about the
previously self-archived preprint as constituting "prior publication."

> Of course, one could break one's word and still post the paper.

Manfredi is again conflating preprint and postprint, Inglefinger and
copyright, past and future. One has ALREADY posted the preprint, before
even submitting the paper to the journal. The only thing one can "break"
in now signing the copyright transfer form that attests to its not having
been previously "published" is some tortured semiotic attempt to construe
preprint self-archiving online as prior publication!

What the author may do with the postprint AFTER signing the copyright
transfer agreement (if he has foolishly, or forcibly, signed an agreement
that transfers all rights to the publisher) is a bit more complicated,
but only a bit. If he has transfered all rights then he may now only
self-archive the corrigenda (and insert the URL of the already archived
preprint, indicating that these changes need to be made to make the
unrefereed preprint conform to the refereed postprint: ).

But in the (anonymous) case under discussion here, the author need
not bother resorting to this preprint-plus-corrigenda strategy, for the
agreement described above explicitly allows the self-archiving of the

> However,
> one would place oneself in the uncomfortable position of being in breach of
> contract (nothing to do with copyright, as Stevan correctly points out).

In "breach of contract" only on an incoherent construal of what constitutes
"prior publication." I wouldn't worry about it. The limitless resources
of hermeneutics are always there to render any contract as putatively
breached if one applies sufficiently anomalous construals of its terms.

> I am no lawyer, but in what way is being in breach of contract "not a
> legal matter" (as claimed in

I stand corrected. I should have said "it is not a copyright matter." It
is a legal matter in precisely the same sense that any abitrary
misconstrual of the terms of a contract would be a legal matter if
that construal could be sustained. Here, this sense of "prior publication"
cannot be sustained.

If the contract instead said:

"I swear that I have not previously self-archived the unrefereed preprint
of this paper on the Web" then it would be a substantive violation of
the contract if one had. But I know of no copyright transfer agreement
that states this arbitrary stipulation in this unequivocal way -- and I
hope authors would have the good sense to strike it out of any contract
before signing it if it did.

> Perhaps the fact that the greatest majority of researchers do NOT
> self-archive may have something to do with this?

I should hope not! Researchers have been sluggish about self-archiving
for many fuzzy reasons, most of them at last showing signs of clearing
up. But I rather doubt that this particular piece of fuzzy reasoning
accounts for a significant proportion of the variance...

> If only things were black and white

They're getting there, they're getting there. Un po piu di pazienzia!

Received on Mon Aug 12 2002 - 00:11:37 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:37 GMT