Re: Elsevier Gives Authors Green Light for Open Access Self-Archiving

From: Stevan Harnad <>
Date: Wed, 16 Jun 2004 15:02:58 +0100

In a press release announced in Peter Suber's Open Access News yesterday
Johns Hopkins University Press announced that it too has given its green
light to self-archiving:

> JHU Press is making revisions to its standard journal
> agreement, clarifying some personal re-use rights. The new
> statement will tell authors:
> You have the following nonexclusive rights: (1) to use the
> Article in your own teaching activities; (2) to publish the
> Article, or permit its publication, as a part of any book
> you may write; (3) to include the Article in your own
> personal or departmental database or on-line site; (4) to
> include the Article in your institutional database provided
> the database does not directly compete with either the
> Johns Hopkins University Press or Project Muse, is non-
> commercial, is institution specific and not a repository
> that is disciplined based and/or accepts contributions from
> outside the institution. For use (4), you agree to request
> prior permission from the Press. For all rights granted in
> this paragraph, you agree to credit the Press as publisher
> and copyright holder.

Some of this may sound a bit confusing to authors, but actually it is
extremely simple. The simple version is: Just look at (3): It gives
the green light for all the self-archiving you need (institutional
self-archiving, which include's the author's own institution-provided
(i) personal website, (ii) departmental website or (iii) institutional
website). Permission is only required for websites at other institutions
(particularly commercial ones that might be competing with the publisher).

I clarify the underlying motivation and legal and technical considerations

The SHERPA/Romeo directory of publishers' self-archiving policies is
maintained at

There is also a Romeo portal at that presents
the self-archiving policies in terms of individual journals rather than
just publishers, in order to help prospective self-archiving authors, their
employers and funders.

I reproduce below a message just sent to SHERPA/Romeo about what policy
details I think should and should not be echoed in the Romeo directory,
and why. Needless to say, it is only the meaningful details that should be
echoed, not the meaningless ones, for that would only be to amplify noise
and create confusion, rather than to clarify self-archiving for authors, their
employers and their funders:

(a) Romeo should not endeavour to restate every nuance of each publisher's
policy. Leave that to the journal's own copyright policy statement
(which is always linked to the Romeo listing). Romeo needs to provide
the relevant broad-stroke data plus the link; it is not an exhaustive
inventory of every clause any publisher ever inserted in any agreement!

(b) This is especially true when the publisher makes detailed distinctions
that are either irrelevant or even meaningless: otherwise Romeo is not
providing information but merely amplifying noise.

(c) There is a specific example of this in the JHUP case: In their (3) they
give the JHUP author the green light

    "(3) to include the Article in your own personal or departmental database
    or on-line site"

That is it: That is all the information that is needed by the prospective
self-archiving author. And it does not require the author to ask for
any further permission. The further distinction, which is meaningless,
that JHUP says requires further permission, is:

    "(4) to include the Article in your institutional database provided
    the database does not directly compete with either the Johns Hopkins
    University Press or Project Muse, is non- commercial, is institution
    specific and not a repository that is disciplined based and/or
    accepts contributions from outside the institution. For use (4),
    you agree to request prior permission from the Press.

I know *exactly* what JHUP's lawyers mean here (and it's silly of them,
but it's ok): They want to make sure that giving their formal green
light to author self-archiving is not construable in court as giving
the green light to 3rd-party rival publishers who then have the right
to publish and sell a competing version, free-riding on the primary
publisher's investment and competing with his product. (1st-party is
the author and his institution/funder, 2nd-party is the publisher,
3rd-party would be anyone else.)

As I say, this is silly (because once the article is available for free
for all on the web, there is no point for a competing publisher to try
to sell it!), but it is ok for JHUP to make this condition if they wish,
because the author's own university, self-archiving its own article
output, is *not* a 3rd-party publisher.

Therefore (4) is irrelevant for institutional self-archiving and
the author need not ask for further permission to self-archive.

The further distinction, however, between (i) the university-provided
disk-sector and format for the author's "personal" or "departmental"
"on-line database" or "on-line server" and (ii) the university-provided
disk sector and format for the author's "institutional database" is
a completely meaningless distinction based on a (lawyerly) technical
failure to understand institutional online servers. Under-informed
JHUP lawyers are here making incoherent analogies with the paper medium,
based on making technically and practically meaningless, hence
untenable, distinctions!

The author's institutional OAI archive, like the author's institutional
OAI home archive, or institutional OAI departmental archive, is indeed
"institution-specific," "non-commercial," and "does not compete with JHUP
and Project Muse" (i.e., it is not a rival 3rd-party publisher re-selling
JHUP products, sanctioned under the guise of "self-archiving"). So clause
(4) is irrelevant in this respect.

There is one condition, however, that *is* technically meaningful,
and hence could be explicitly echoed in Romeo, and this pertains to
central, discipline-based archives: Again, this stipulation is silly,
because central archives like Arxiv and CogPrints are not only every
bit as non-commercial, and every bit *not* competing publishers, as the
author's own institutional archive is. But, in their lawyers' concern to
make sure they are not formally sanctioning 3rd-party publishing
in giving their green light to author/institution self-archiving, some
publishers have formally ruled out central, discipline-based archiving
as a precaution:

Central archives are *not* just an institutional disk-sector, and they
*do* accept articles from other institutions, hence they do not quite
fit the technical meaning of "self" in 1st-party "self-archiving"
if the "self" is construed as the author himself, archiving his own
paper in his "own" archive:

The author's institutional disk sector for his homepage, departmental
page, or institutional page are all clearly 1st-party "own" and
"self". But a central archive, at another institution's website, is not.

To repeat, this distinction (which could easily be finessed technically
by distributing the central archives through local institutional nodes
instead of localizing them) is silly in the case of discipline-based
central archives like Arxiv and CogPrints, which are clearly not rival
publishers but simply another way of achieving *exactly* the same outcome
as putting one's paper in one's own institutional archive: In either case,
the result is a free article, OAI-compliant, openly accessible webwide,
its metadata harvestable by google, OAIster, and others, etc.

But the JHUP lawyers feel more comfortable adding this extra explicit
restriction so that a true rival 3rd-party publisher cannot use JHUP's
official green light for author/institution self-archiving in order to
legally justify republishing and re-selling JHUP products, as a free-riding
3rd party.

So the condition is harmless (and specific exemptions can no doubt
be negotiated by the author with JHUP for OAI archives like Arxiv or
CogPrints, if the author wishes, either before self-archiving, or,
more sensibly, after, if JHUP ever decides to ask him to remove his
self-archived article -- as no publisher has ever done, for any of the
260,000 articles self-archived in Arxiv since 1991!).

So, please, just list condition (3) in SHERPA/Romeo, using only the
wording "On author or institutional server only". That covers it as
closely as Romeo should get involved in it. Don't re-iterate incoherent
details. You might add only that self-archiving elsewhere than in the author's
own institutional archives requires permission from JHUP.

I would also request, again, that SHERPA/Romeo drop 2 out of the 3
redundant and misleading alternatives provided as "examples" in the
"cut and paste" section. It should not be Romeo's role to encourage
publishers to make meaningless distinctions! Please drop the 1st and 3rd
of the following from your list on

--> # On author's or employer's web site only
    # On author or institutional server only
--> # On author's personal web site

Finally, I hope it is obvious that the JHUP stipulation about "not a
repository that is disciplined based" is merely another reference to
3rd-party central archives, and should not be construed as contradicting
the JHUP's own green light for "departmental" archives (which are,
of course, also "discipline based")!

Lawyers! Love 'em, but they *do* leave you exhausted...

Stevan Harnad

    Pertinent Prior AmSci Topic Threads:

    "Open Letter to Philip Campbell, Editor, Nature"

    "Re: Central versus institutional self-archiving"

    "Re: Nature's vs. Science's Embargo Policy"

    "Re: Free Access vs. Open Access"

    "Re: Legal ways around copyright for one's own giveaway texts"

    "Re: The Special Case of Law Reviews"

    "Re: Elsevier Gives Authors Green Light for Open Access Self-Archiving"

    "New JISC-funded Rights Metadata for Open archiving (RoMEO) project"

    "Re: Project RoMEO Listing of Journal Copyright Agreements"

    "DOAJ, OAIster and Romeo should chart growth, as EPrints does"

    "Please compare Romeo functionality and provide feedback"

NOTE: A complete archive of the ongoing discussion of providing open
access to the peer-reviewed research literature online (1998-2004)
is available at the American Scientist Open Access Forum:
        To join the Forum:
        Post discussion to:
        Hypermail Archive:

Unified Dual Open-Access-Provision Policy:
    BOAI-2 ("gold"): Publish your article in a suitable open-access
            journal whenever one exists.
    BOAI-1 ("green"): Otherwise, publish your article in a suitable
            toll-access journal and also self-archive it.
Received on Wed Jun 16 2004 - 15:02:58 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:47:29 GMT