UK Select Committee Inquiry into Scientific Publication

Stevan Harnad harnad at ecs.soton.ac.uk
Fri May 7 06:56:32 EST 2004

On Thu, 6 May 2004, Richard Poynder wrote:

>sh> it is OA and OA publishing that are conflated (i.e.,
>sh> treated as if they were the same thing). OA self-archiving is thereby
>sh> *overlooked* as a means of providing OA, let alone the far 
>sh> faster, surer means that it in reality is, and the means that is already
>sh> providing far more OA -- and could easily provide it all: 100%
>sh> The second conflation, as I said (after the conflation of OA with OA 
>sh> publishing, overlooking OA self-archiving) is the conflation of the 
>sh> journal-affordability problem with the access/impact problem: 
>sh> And you are doing it again!
> There are a number of reasons why people conflate things. One is that
> the person guilty of the conflation does not understand the difference
> between the things they are conflating. Another is that although they do
> understand there are differences they do not believe those differences
> to be significant enough to treat them as entirely separate things. 

In this case, the two amount to the same thing: To understand the
difference is to understand *why* they should not be conflated: Conflate
them and you overlook *actual* OA (20% from OA self-archiving vs. 5% from
OA journal publishing) and, more important, you lose immediate *potential*
OA (at least 83% from OA self-archiving). 

Whoever thinks that is not significant surely does not think OA is
significant (and hence should preface all pronouncements on OA by saying
they do not consider it significant).


What Richard seems to keep doing by way of reply to this point is
to pass in silence over the objective facts raised, and to raise instead
subjective speculations: In the prior iterations these were speculations
about a hypothetical future cancellation catastrophe point, should OA
self-archiving grow; this time they are speculations about a hypothetical
future change-of-mind by the green publishers, should self-archiving grow!

I recommend a moratorium on paralytic speculation, and an acceleration
on practical self-archiving in its place!


> The fact that so much conflation takes place around the topic
> of OA suggests that... the differences between OA
> and OA publishing, and between the journal-affordability problem and the
> access/impact problem, are (to some people) not seen to be significant
> enough to keep making the distinction. 

No doubt! But to those people for whom OA itself is significant, this
distinction is necessarily significant. To those for whom only journal
affordability is significant, I respectfully suggest that both they and
OA would be far better off if each hoed their rows separately. The need of
immediate relief for the journal budget problem will only get in the
way of the need of immediate relief for the research impact problem,
and vice versa.

The same is true for those who wish to restructure journal publication:
a worthwhile agenda, but not if it obscures or is at odds with the wish
to maximize research access, now, by making articles OA, now, rather
than by waiting for a possible restructuring of journal publication some
day, some way. Let these rows too be hoed separately, rather than being
conflated, obstructing one another.

> If self-archiving is being overlooked, by the way, it is perhaps - as I
> suggested previously - because although it has the merit of providing
> a fast-track route to OA, it is not a long-term solution to anything
> in itself, since it is parasitical on (today) TA publishers. 

(1) Self-archiving is most certainly not being overlooked by the
many authors who are doing it: At least 3 times more authors (20%)
are self-archiving their articles today than are publishing them in OA
journals (5%); and self-archiving is growing faster too. What is being
overlooked is the far greater *potential* of self-archiving to provide
dramatically more immediate OA than it already provides so far: indeed,
83-100% of the entire target corpus! It is this profoundly underutilized
power that is being overlooked -- by too many those who purport to be
OA proponents and activists! And this is happening at least in part
directly because of the 2 conflations I mentioned.

Maybe some of the ostensible OA proponents, once they deconflate, will
realize that they are not OA activists at all, but simple activists
for easing the library serials budget, or activists for resturcturing
the journal publishing system. That is all fine, but then that should be
realized, and declared as such, so measures aimed primarily at easing
serials budgets or restructuring journal publication don't keep getting
represented as OA, obscuring other measures that really are OA, primarily
or solely OA, already generating a great deal more OA than these other
measures, and, most important, *capable* of generating 83-100% immediate

(2) You say self-archiving is "not a long-term solution to anything
in itself," but you are just conflating again! Self-archiving is
not a solution (short- or long-term) to the journal pricing problem, 
but it is not primarily intended to be. Nor is it a solution (short-
or long-term) to the problem of restructuring journal publishing, or
of converting journals from the TA to the OA cost-recovery model. It
is merely a solution to the problem of providing immediate OA! 

Is self-archiving a long-term solution for providing OA? Well, the
articles self-archived in the Physics Arxiv in 1991 are still alive
and well and OA today, thank you very much, and as heavily used as need
dictates, a dozen years later! If the physicists had instead fretted
needlessly about a "long-term solution" instead of immediately
self-archiving at that time, they would have needlessly lost another
dozen years' worth of usage and impact, as those who did not self-archive
at that time have done.

Yet it is precisely more of this pre-emptive waiting and worrying that
Richard appears to be recommending to us (exhibiting at least 2 of the
31 familiar symptoms -- all treatable! -- of Zeno's Paralysis).


*All* research article usage is "parasitical" on today's journal
publishers (whether TA publishers or OA publishers). One of the uses
to which authors put their articles in the online age is to make them
accessible online to any would-be users whose institutions cannot
afford the toll-access to the TA version. So far, that growing practise
has co-existed with the TA publishing system without inducing any
cancellation pressure, even in fields where it has reached 100%. If the
TA journal publishing system should ever have to adapt to the growing OA
provision in some way, it can and will. But there is no earthly reason
for authors to go on needlessly sacrificing their potential impact
waiting for some *pre-adaptive* change in publishing to occur first!

> [Self-archiving] is
> a street protest able to gain the attention of the powers-that-be -
> and with luck encourage them to seek a solution to the problems of (a)
> the journal-affordability problem and (b) the access/impact problem -
> but it is not a government in waiting.

Nothing of the sort (unless we consent to conflate)! Self-archiving is
not a "protest" at all, nor a means to some other end! It provides an
immediate, face-valid long-term solution to the only problem that it
was intended to remedy: the stanching of all further needless impact
loss, by providing Open Access to one's own research articles, thereby
maximizing their accessibility, usage and impact -- both immediately and in
the long-term.

> The reasons for treating the journal-affordability problem and the
> access/impact problem as though they were one problem are not without
> merit [sic]. 


> Those TA publishers who have allowed self-archiving have taken
> the dual gamble that (a) not enough researchers will self-archive to
> make a difference (and that if they do the search tools for utilising
> the self-archived material will not be competent enough to compete
> with their own search tools) and (b) that they can continue to keep
> selling their journals to libraries at the same level/maintain their
> high profit margins. 

Not content with second-guessing the future course of publishing,
Richard now second-guesses publishers' motives (and, below, their
future actions).

Why all this speculation? The only thing at issue (for the researcher) is
daily/monthly/yearly research usage and impact that are being needlessly
lost, and can be immediately remedied by self-archiving. Richard seems
to suggest that researchers should just keep on losing that impact,
because their lost impact is not the real problem: The real problem
is the library's serials budget crisis and the need to restructure the
journal publication system (and the author self-archiving is merely a
"protest" against this state of affairs!).

May I suggest a more benign interpretation of why 83% of journals have
already given the green light to self-archiving? Because (i) they have
recognized the sizeable benefits of OA for their authors and for research,
(ii) they are not prepared to make the sacrifice and take the risk of
converting to OA, but (iii) they do not want to stand it the way --
or to be perceived as standing in the way -- of authors providing OA
for their own articles if they wish, in order to derive those sizeable
benefits for their own research, and for research progress as a whole.

Some publishers may be very aggressive in the market place, in their
pricing policy, but that is another matter, and another problem. And the
two should not be conflated. A green light for OA self-archiving is a
green light for OA, and should be appreciated as such -- and acted
upon! It has nothing to do with whether or not the journal is over-priced.

[Just as there were those who counselled paralysis because of what
publishers might think or do if we self-archive -- ignoring a dozen years
of peaceful precedent -- there are now those who counsel paralysis even
when we have the publishers' green-light, because if we go ahead and
act upon the green light, that will make the light turn red (or gray,
rather)! If these are the "merits" of conflation, I'm for renouncing
such merits, and conflation!]

> Let's leave aside the most obvious interactive relationship between (a)
> and (b) and assume that publishers' gamble over (a) is failing and, with
> libraries cancelling journals in ever greater numbers, and bargaining
> over the cost of electronic bundles in an ever more aggressive way,
> that the gamble over (b) is failing too. 

Let us, in other words, assume that Richard's speculation comes true!
Fine. I think it is completely irrelevant to what the research community
should do now, but fine, it is one of the logical possibilities, so let
us suppose it does gradually come to pass eventually, in the aftermath
of 100% OA. So?

> While self-archiving may be
> driven by the access/impact problem and journal cancellations by the
> journal-affordability problem, the response to the failure of their gamble
> by TA publishers - regardless of whether it is a response related to (a),
> or a response related to (b) - is likely to impact both the access/impact
> problem (a) and the journal-affordability problem (b). 

This was a rather complicated mouthful, but I take it that Richard means
that if approaching 100% OA from self-archiving results in increased
cancellation pressure then it will result in increased cancellation
pressure, and the journal publication system will have to adapt to it --
probably by cost-cutting and downsizing to the essentials; possibly
eventually by recovering these costs by converting to the OA journal

So these are some possibilities about the way the journal publishing
system may evolve in the OA era. But what is the implication of these
possibilities right now?

> For instance, in response to the failure to continue selling journal
> subscriptions at the same level publishers may harden their attitude
> to self-archiving, on the grounds that journal cancellation is being
> facilitated by self-archiving. This suggests that both problems are
> sufficiently linked to treat them as one problem and that a common
> solution is called for.

So authors should not self-archive (or, equivalently, self-archiving
fails to solve "the problem") because the green publishers might some
day change their minds and go gray again? 

It is clear that people who think this way are not researchers: Don't
bother immediately beginning to maximise your research impact, because
one day publishers may decide to withdraw their green light?

Has Richard ever contemplated the possibility that OA provision has
absolutely *nothing to do with publishers*? that is is not a "protest,"
not a means, but an end in itself? and that the green light is merely
one of the ways to energize those researchers who did not have the good
sense to just get on with self-archiving, as hundreds of thousands of
researchers have been doing ever since 1991, without giving publisher
self-archiving policy a thought?

The conflation of the journal pricing problem with the access/impact
problem, and the preference for speculating and second-guessing over doing
the optimal and the obvious, at once, leads some to see non-solutions
where others see immediate, dramatic, revolutionary and irreversible
research progress:


Can we agree that Richard, and anyone who is of a like mind, should go
ahead and keep seeking common solutions for the unitary problem they
have in mind (whatever it is), while in the meanwhile we researchers go
ahead and solve our own (separate) problem, which is to maximise the
impact of our own articles by providing OA to them for all those would-be
users whose institutions cannot afford the access-tolls for the journal
in which they appear -- by self-archiving them?

And I won't say that your solution to your problem (whatever it is) is not
really a solution, because it is not a solution to *my* problem (which
is to make my articles accessible to all would-be users immediately),
if you do the same for me. In other words, let us agree not to conflate
one another's problems!

>sh> I am a hard-pressed librarian. Every year, I can afford fewer and fewer
>sh> journals because of their price-rises. So I have to decide what to cancel.
>sh> Normally, I consult my user logs and users, about which journals they use 
>sh> and need most, and I cancel according to a multiple constraint 
>sh> satisfaction analysis of the costs/benefits (including journal impact 
>sh> factors).
>sh> I also have a table of the "green" journals -- the 83% that have given
>sh> their green light to self-archiving. 
>sh> So consider the following scenarios...
> But if the number of journals libraries can afford to subscribe to is
> falling, and libraries are more and more resisting price increases to
> electronic access (i.e. the journal-affordability problem) TA publishers
> are likely to conclude that they can no longer achieve the level of
> profits their shareholders demand. Regardless of what individual journal
> cancellation decisions libraries take, therefore, the overall market for
> TA publishers (given that TA publishers generally own very large bundles
> of journals, not a handful) is becoming less and less attractive. If this
> leads them to exit the market there will be inevitable repercussions for
> the access/impact problem. Apart from anything else it could lead to a
> sudden closure of some of the journals that researchers need in order
> to make their impact. Again, the journal-affordability problem and the
> access/impact problem are intertwined.

We are back to the catastrophe hypothesis: What exactly is Richard
recommending? That researchers refrain from maximizing their impact
now, even though it is feasible (and even though it even has the green
light from 83% of journals) because of the speculation that it *might*
some day lead to a catastrophe? For all I am trying to do is to get
all researchers to self-archive, so we can have OA, at last. If this
open, green road is sacrificed in favor of waiting for some other
catastrophe-avoiding road (which? how long?), it sounds very much as if
researchers are being invited to forego their impact on the strength of
a gloomy and irrelevant hunch...

"Sudden closure" of journals? What do you imagine journals are? They are
titles, consisting of names, editorial boards and authorships. If a
publisher cannot or no longer wishes to produce a still-viable title,
the title (and editorial board and authorships) migrate to another
publisher. As I said in my prior reply, there would be plenty of lean
and hungry publishers ready to take over viable titles if their publisher
discard them, for whatever reason.

>sh> MPs may be professional puzzlers, but it is the research community that 
>sh> can and must act. I had hoped that parliament, as the purveyor of the 
>sh> research purse-strings, might help persuade researchers to provide OA. 
> Would it be correct to say that you believe the only role for government
> here is to insist that the output from publicly funded research should be
> OA? (A suggestion, by the way, made by both Harold Varmus and Fred Friend to
> the UK Select Committee - and so perhaps  a proposal that all OA advocates
> could agree on?).



Stevan Harnad

More information about the Jrnlnote mailing list

Send comments to us at biosci-help [At] net.bio.net