Re: STM Talk: Open Access by Peaceful Evolution

From: Stevan Harnad <>
Date: Mon, 24 Feb 2003 15:50:36 +0000

On Mon, 24 Feb 2003, Jean-Claude Guédon wrote:

>s> The lament is: "We are buying back what we give away." I said this was
>s> incorrect. For books it is moot, because they are not given away by
>s> authors but written for royalty. Refereed research is indeed given away,
>s> but it is not the refereed research output of their own institution that
>s> a library is buying in, but the refereed research output of all other
>s> institutions. The lament cannot be read as a collective macro-statement
>s> by all libraries, because "we" and "our" completely lose their sense if
>s> construed collectively. Yes, there is a collective resolution to this
>s> anomaly, but it is the institutional self-archiving of each institution's
>s> own refereed research output:
>J> Why do "we" and "our" completely lose their sense? I do not understand
>J> your objection.

For the very same reason that it does not make sense for books:
Libraries are not buying back their own institutional book output, they
are buying in the book output of other institutions. It does not help to
say "collectively, we are buying back our own collective output," for
this completely conflates who is producer and who is consumer in each
particular case (and this is without even mentioning that it is the
authors who are doing the producing and that books are not author
give-aways). One might as well say, of the stock-market, that "we are
all buying back our own stocks." No, there are buyers and sellers; there
is no collective "we" that is both the giver and the receiver.

But the main error in putting it in this way is of course with refereed
research; for whereas books are not an author give-away, refereed
research is. That still does not change the fact that the buy-in, from
an institution's point of view, is not a buy-back of its *own* give-away
research output, but a buy-in of *other* institutions' give-away
research output.

And the remedy for this is also quite clear, and does not lie with the
publishers, who cannot be expected to do it for us. If, unlike books,
refereed research output is indeed an author/institution give-away,
written only for maximal research impact, let the author/institutions
take the natural step to remedy the problem of access-denial for those
potential users whose institutions cannot afford the access-tolls,
according to the Golden Rule: "Self-Archive Unto Others As Ye Would Have
Them Self-Archive Unto You." It is *here* the the "we" is indeed a
collective one, not in the economics of library serial acquisitions.

>J> My response is that researchers should give away *access* to
>J> their publication (i.e. for free) to their colleagues, retain their
>J> copyright and give or - why not - sell their texts (non exclusively)
>J> to commercial interest if they feel it can enhance the visibility
>J> of their ware.

At the present moment, if researchers want their papers to be
peer-reviewed and certified as such by an established journal with a name
and track-record -- as virtually all researchers do -- they must (and
should) submit it to a refereed journal. Anything predicated on their
doing something *other* than that is not only unrealistic, but probably
not good advice to give to researchers! Most refereed journals today
are still toll-access journals. So that would be the end of the story,
insofar as open access is concerned, wherever no suitable open-access
refereed journal like BMC exists -- if it were not for the self-archiving
option, which is intended as a *supplement* to -- not a *substitute*
for -- publishing in the refereed journals. (Open-access, let us not
keep forgetting, is about open access to *refereed* research -- not to
unrefereed research, later to be sold on some hypothetical market!)

So, we agree on open-access. I think we agree on peer-review. We
disagree on the *necessity* of copyright retention (though we agree on
its desirability). Self-archiving is a tried, tested and proven road
to immediate open-access. That is undeniable. My only point was that
this is entirely in the hands of the research community, not the library
community, nor the publisher community. (And the "we give it away and
buy it back" dictum, though related to the problem and its solution,
fails to capture the underlying causal dynamics.)

> That more and more journals are responding more positively to the idea of
> self-archiving is true; it may well be part of their fear of alienating
> researchers too much.

Maybe, but I see no reason to look at it so uncharitably. I don't think
journal publishers are villains. I think they, as much as authors, are
victims of the paper-centred system they have inherited. Right now, this
paper-centred system has been quite remunerative for some (but by no
means all, or even most) refereed journal publishers. But the online age
has changed things radically, and the new possibilities are only now being
realized. For researchers, one new possibility -- maximising the impact
of their research output by self-archiving it -- is immediate, and has
already been enjoyed by tens of thousands of authors of preprints and
postprints of refereed research publications. For publishers, there have
not yet been any consequences, and it is conceivable (though unlikely)
that there never will be, with the toll-access versions continuing to
be used by the researchers at the institutions that can afford it, and
the self-archived open-access versions used by the researchers at the
institutions that cannot afford the tolls. Or there may be a gradual
downsizing of refereed journal publication to only the essentials
(mainly peer review implementation) with the paper and composition
jettisoned and all distribution, archiving, and access-provision offloaded
onto the institutional eprint archive network.

So the reason more and more refereed journal publishers are supporting
self-archiving, despite the element of risk for them, is that it is so
obviously beneficial to research and researchers.

> But when they do not, the fear of scholars is real, the
> self-censorhip also. And, moreover, in many countries, your idea of
> self-archiving the pre-refereed version plus the corrigenda would not hold in
> court, so far as I understand copyright and authors' rights, plus Bern, etc.

Intuitions differ on this question, but here are the facts: Since the
early 90's, hundreds of thousands of peer-reviewed papers have been
self-archived without a single court challenge. Moreover, talk of a
"court challenge" is rather alarmist, as what would really happen,
if anything ever did, would merely be that an author receives a
notice from the legal department of the journal requesting that he
remove a particular self-archived article. At that point the author would
have the choice of complying with the request and removing the file in
question. That is all that is at issue. And it has not even happened once
in ten years of growing self-archiving -- nor is it likely to happen,
for during that decade publisher support for self-archiving was likewise
growing, because of its obvious benefits to research and researchers.

It is also a fact that despite a decade of (1) growing self-archiving,
and (2) growing publisher support for self-archiving, and (3) no authors
asked to remove self-archived papers -- I doubt the number is zero,
but it is clearly vanishingly small: have you even heard of a single
case? Please don't reply with cases other than the self-archiving of
refereed research! -- there are nevertheless many researchers whose
self-archiving is inhibited by groundless legal worries:

My question to you, Jean-Claude, is this: In view of the empirical
evidence to date, is it our role to amplify these groundless worries, by
adding our own speculations about hypothetical "court challenges"? Or
should we be encouraging the growing numbers of self-archivers to grow

> Plagiarism rules will most certainly apply and they too are part of the legal
> equation.

Plagiarism is completely irrelevant to the self-archiving of refereed
    "1.3. Distinguish between copyright protection from
    theft-of-authorship (plagiarism) and copyright protection from
    theft-of-text (piracy)"

> In any case, so long as you do not have a clear court case in hand
> in a significant country, preferably the US, your argument remains largely
> moot. I would love to believe it; I simply find myself unable to although I
> am sympathetic to our common cause (more than sympathetic, actually, as you
> well know) and wish it were true.

I do not understand your logic. Hundreds of thousands of peer-reviewed
papers have been self-archived, their number is growing, the number of
publishers supporting self-archiving is growing (half the journals covered
by the Romeo Table above already support it), and you are unsure until
and unless there is a court case? (The "preprints+corrigenda strategy"
is meant as a reductio ad absurdum for the hesitant, not as a hypothesis
about future legal contingencies!)

Is that meant to be your advice to the self-archivers and would-be
self-archivers: "Wait for a court challenge?" I suggest that that would
be very bad advice! And if that is not meant to be your advice to
self-archivers, then why are we even discussing it here, rather than
ways to accelerate the volume and rate of self-archiving?

> OAIster, so far as I know, has not published any statistics on the status or
> age of the contributors.

But Dear Jean-Claude, neither have you! On whom is the burden to show
that "1,093,169 records from 144 institutions" are primarily from
oldsters: Oaister's or yours, as you are the one who claims it is so!

> On the other hand, remember that if PLOS came to nought in its first
> incarnation, it was, I believe, largely because young scientists,
> despite all their anger, could not face taking so many risks.

I agree that the PLOS signatories' threat to publishers not to publish
in their journals unless they become open-access journals was a rather
quixotic one (though I would still like to see your age statistics!).
But it only confirms that the publishers are not the right target. The
research community should target *itself* if it wants open access to its
refereed research output, and that is precisely what the growing numbers
of self-archivers are doing. Open-access journals are welcome too, but
whereas open-access to all 2,000,000 annual articles in the planet's
20,000 refereed journals could come virtually overnight through
self-archiving, it would take considerably more nights to create or
convert 20,000 open-access journals. Petition signatories cannot do the
latter either; but they can certainly do the former.

> Everyone agrees that the physics community (and not even all of physics
> - high energy began earlier than dense matter, for example, for reasons
> that, incidentally, I have never seen explained in the literature - anyone
> knows otherwise? ) does not quite behave like most other disciplines.

Yes, they will get historic credit for having been the first to discover
the benefits of self-archiving. But those benefits (maximized research
visibility, accessibility, and impact) are certainly in no way unique or
peculiar to physics! Physicists simply realized it earlier.

> Mathematicians and astronomers follow suit; molecular biologists and
> chemists don't. In physics, not only are papers multi-authored (which
> happens also in many other disciplines), papers are vetted through
> intense lab discussions (which also happen in several disciplines)
> and research instruments are often so very rare and costly (which is
> found in a few other disciplines such as oceanography and astronomy)
> that performing the experiment ensures ownership of the terrain for a
> little while. It happens that physics has all three characteristics.
> Why should mathematicians should behave as they do remains a little
> unclear to me.

Future historians will no doubt analyze the time-course of the growth of
self-archiving and open access. Your hypotheses are interesting (though
at least one of them is certainly wrong: the leaders were not and are
not the experimentalists, but the theorists in physics; experimental
physics still lags behind), though they seem to focus a little too much
on the self-archiving of pre-refereeing preprints, whereas the real
target is the refereed postprints, on which the above seems to have
little bearing. In addition, your hypotheses now have to be extended to
cover computer science, with 500,000 self-archived papers harvested by

What ResearchIndex demonstrated was that centralized,
discipline-specific archives like the Physics ArXiv are by now merely
the visible (because centralized) tip of the iceberg, and that
self-archiving has been growing anarchically and dramatically
on researchers' own websites in many (probably all) other
disciplines. OAister's "1,093,169 records from 144 institutions" -- not
just the 200,000 from the Physics ArXiv -- are further evidence for this
(though not all of OAIster is refereed research preprints and postprints).

The fine-grained analysis will come from the historians and

> This said, if all journals unambiguously agreed that self-archiving is fine
> and solemnly declared that it in no way affects publishing in the
> corresponding journals, then I might begin to agree with you. We are far from
> that. You, yourself, are spending a fair amount of time trying to clarify
> issues with publishers, such as Nature, most recently. In effect, you are
> trying to clear the way to help convince the hesitating lot to move ahead.
> But if they hesitate, it is not just because they do not see things clearly
> (or your way), it is because they face real obstacles, locally or nationally.

I and others are trying to help accelerate self-archiving by clearing
away the obstacles (mostly imaginary) that are slowing those who do not
yet self-archive. If the growing number of self-archivers in the
past decade, or the physicists who first realized its feasibility and
potential, had instead waited till "all journals unambiguously agreed
that self-archiving is fine and solemnly declared that", then a lot of
research progress would have been lost.

We are indeed "trying to clear the way to help convince the hesitating
lot to move ahead," and worry about legality is only one of the many
groundless worries that we must work to remedy: There are at least 25
more! See the "I-worry-about..." FAQs at:

1. Preservation:
2. Authentication:
3. Corruption:
4. Navigation (info-glut):
5. Certification:
6. Evaluation:
7. Peer review:
8. Paying the piper:
9. Downsizing:
10. Copyright:
11. Plagiarism:
12. Priority:
13. Censorship:
14. Capitalism:
15. Readability:
16. Graphics:
17. Publishers' future:
18. Libraries'/Librarians' future:
19. Learned Societies' future:
20. University conspiracy:
21. Serendipity:
22. Tenure/Promotion:
23. Version control:
24. Napster:
25. Mark-up:
26. Classification:

>s> I am afraid I have to repeat that this is not at all relevant to what
>s> it is that I am talking about. Let editors/gatekeepers keep doing
>s> whatever they are doing; we are not talking about that. We are talking
>s> about providing open access to the outcome of their gate-keeping system
>s> by self-archiving. (We must not confuse gate-keeping with the toll-gating!)
>J> The point is that locally influential scientists that have advantages
>J> of all kinds in being gatekeepers and remaining in that position
>J> will or at least can use that power to influence younger, more
>J> vulnerable colleagues.

Influence them to do what, insofar as self-archiving is concerned?

>J> For that reason, the scientific community is
>J> divided. I tried to say that, roughly, it was divided between the
>J> powerful cynics, the vulnerable and the idealists. If you say that
>J> this bears in no way on the problem, it confirms my suspicion that
>J> you want reality to bend to your (superb) logic, rather than the
>J> other way around.

No, it is genuine befuddlement! I haven't any idea what the causal link
is between editors/gatekeepers, their undeniable influence on the young
(who submit to their journal) and self-archiving: Is this merely a
repetition of the publisher/legality worry, but this time substituting
the journal's editor for the journal's publisher as the villain?

>s> What do tenure/promoting mechanisms etc. have to do with the virtue of
>s> self-archiving one's own refereed research output -- apart from the fact
>s> that they will certainly reward the additional research impact
>s> self-archiving brings?
> >
>J> They certainly won't "certainly reward the additional research impact
>J> self-archiving brings". What allows you to assert this thesis with such
>J> assurance?

I am so happy you asked me that! I can assert it with such assurance
because it is a tautology: They *already* reward the magnitude of
research impact, hence, whatever *increases* the magnitude of research
impact will merely draw upon that already existing contingency!

> my feeling is that open-access, including with
> self-archiving, is not going to proceed well so long as a significant
> fraction of researchers actively support the existing system at the exclusion
> of all others. I have met the enemy and it is part of us to paraphrase a
> well-known Pogo quip. This is why I pursue such "distracting" battles. This
> said, I value your clear vision as it often helps me choose my battles, but I
> have come to the conclusion that I must wage a few more than you.

The enemy is not the publishers, not peer-review, not the
tenure/promotion system, not copyright law: The enemy is just the
slowness of the research community in realizing and utilizing
the new possibility opened up for them by the online medium; but
the tempo is picking up, so let us not be so pessimistic!

> scholarly publishing is not a
> logical game and logic only takes you so far. Then comes the messy world of
> alliances, politics, battles and tactics.

The only battle I want to fight is the one to make the token drop in
researchers' heads. Logic may not be enough, but it's sure better to
have it on our side rather than elsewhere!

>s> it was a librarian, Ann Okerson, who first gave me the idea (which
>s> I at first rejected as an impossible, indeed unrealistic dream)
>s> of open access when I was still obsessed merely with online access!
>J> We fully agree on open-access, but have you considered the possibility
>J> that you are a little obsessed again, this time by self-archiving? :-)

But online-access (the old obsession) *did* prevail, completely! So now it
is time for open-access to prevail. I recognise and value the second
road too (BOAI-2, creating and converting open-access journals), but I
am pretty certain that BOAI-1, self-archiving, is the faster road
(despite Pogo's sluggishness!). And I am also pretty sure that
shadow-boxing with publishers is no road at all.

> I know you think BOAI-1 better than 2 precisely because of the sociological
> homogeneity (just researchers) that you ascribe to it; my objection is simply
> that even BOAI-1 cannot escape the messy world of a complex technical system
> involving a good many varied actors, such as researchers, of course, but also
> peers in all colours and stripes (tenure, granting agencies, publishing,
> etc...), administrators, librarians and, of course, publishers, also of all
> colours ands stripes. In short, where you argue for logical clarity of vision
> (and again, this is very useful, in my opinion: I know you have helped me in
> this regard), I argue for that plus sociological and historical
> sophistication.

But most of those other actors (apart from tenure/promotion/funding
mechanisms, which are implicitly with us, because they are
research-impact-driven) are simply not relevant. Immediate open-access
through self-archiving is a matter that is completely within the hands
of the research community, exactly as it has been for all those
researchers who have self-archived already.

>s> OAIster reports "1,089,937 records from 142 institutions" but that is an
>s> overestimate, because some of their records are not full-text and many
>s> are not refereed research. But there is at least Arxiv (200,000),
>s> plus CogPrints and the other Eprints Archives (19,000).
>s> Then there is ResearchIndex, not yet OAI-compliant, and not all refereed
>s> research either, but 500,000 and the "gray
>s> iceberg" consisting of the hundreds of thousands of refereed papers that
>s> have been self-archived on authors' own websites (but not yet countable and
>s> searchable because not yet OAI-compliant).
>J> You have no proof that, in the archives you mention, the
>J> representation of young professors and researchers is adequate;
>J> neither do you anything about their working environment and
>J> circumstances. In other words, your statistics are too macro to
>J> lend themselves to a useful interpretation.

I think the burden of proof that there is something *anomalous* about
those who have self-archived so far is on those who wish to claim that
is the case, not on those who do not make that claim, and point out
instead that maximizing research impact by maximizing research access is
is a value that would (and will) be sought by *every* researcher, in every
discipline, once it is known and understood.

Stevan Harnad
Received on Mon Feb 24 2003 - 15:50:36 GMT

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:46:52 GMT