Re: Science 4 September on Copyright

From: Albert Henderson <> <harnad_at_COGSCI.SOTON.AC.UK>
Date: Thu, 17 Sep 1998 13:24:27 -0400

On 15 Sep 1998 Arthur Smith <apsmith_at_APS.ORG> wrote:

> Finally some real numbers...

> Harnad asks also about growth. With doubling every decade,
> the number of published articles probably grew
> a factor of 10 or more (probably significantly more because
> of the rise of the biological sciences) between 1960 and 1995.

Some areas surely grow at this rate. My best estimate of growth of the
numbers of articles, based largely on database production records
collected by NFAIS and the observations of Derek de Solla Price, would
be a growth just over 5X during 1960-95. US academic R&D grew at a
faster rate 1960-1970 and slowed for a year when Congress cut defense
research. Then it picked up again, putting it substantially ahead of
world research for the full period.

Many databases have been forced to cap their coverage during the last
years of this period, excluding many published articles.
This unfortunately interferes with the accuracy of future statistics
as a measure (as well as the usefulness of resource sharing).

In the 1960s, as a result of the embarrassment of Sputnik, support for
US academic R&D AND libraries surged. Libraries also had the benefit of
the Higher Education Act of 1965 Title II-A (college library materials)
which focused on collection development for a few years.

The period 1970 to 1995 indicates growth of US academic R&D by a factor
of 2.5, world research 3.2 and the 41 ARL libraries 1.6. Library
photocopying, particularly interlibrary borrowing increased
substantially -- 132% 1986-1997. Commercial document delivery is not so
well tracked as interlibrary borrowing, but we know that foreign
sources now supply hundreds of thousands of photocopies to US
libraries. This suggests that US sources have dried up.

> It's well known that the serials
> crisis has been brought on by the huge growth in the number of scientific
> articles published - now we have a second reason (a drop in library
> funding) - the problem is laid squarely back at the door of the researchers> and their institutions!

In contrast to the spin put out by universities, attempting to
shift the blame to publishers, I would say the serials crisis was
instigated by cutting library growth. Price noted that libraries and
technical publications grew at the same exponential rate, doubling
roughly every 15 years. (SCIENCE SINCE BABYLON 1961 enl ed 1975 p.
173) That growth has slowed very about half the rate while research
proceeds as before. Here is an extension of the statistics on the
average growth of ten 100+ years old library collections (collected by
F Rider and cited by Price) with new data published by Association of
Research Libraries.

    1938 1.2 million vols.
    1954 not available
    1968 2.7 " "
    1983 4.1 " "
    1997 5.7 " "

Compare the latter figure with 19.2 million vols. projected if pre-1938
growth, keeping up with the work product of world research activity,
had continued. By way of reference, Library of Congress presently
reports 24, Harvard 13.6, Yale 10, CISTI 8, NAL 2.3, NLM 2.2, and
Smithsonian 1.2 million vols.

> Now in the next 10 years, if we see another 60-70 percent drop in
> per-article publication costs, what will libraries do with the savings
> (if any this time)?

The "savings" goes to administration, not the library; it feeds
administrative "bloat." I am told that the indirect cost payments for
libraries, made in connection with Federal research grants, never get
into the hands of the librarian.

And what will researchers do? If there are no libraries, no databases
and no journals, we return to pre-1665 chaos of all formal
communications being nailed to the post, now on the WWW rather than the

As it is, the libraries are half-empty, collecting half (or less
according to studies of monographs) of research published in the last
30 years or so.

US libraries have been more likely to retain journals that their
faculty are associated with. Citation studies in SCIENCE AND
indicate US scientists and engineers cite US authors at an
extraordinarily high rate compared with foreign authors. This suggests
to me insularity that may well be traced to the decimation of library
collections and the absence of foreign authors.

Is there a policy of dissemination? I don't think so. The patterns of
behavior suggest to me that the plan is to ditch libraries and shift
the burden of communications to authors. Unless associations like APS
and federal science agencies exercise their influence on policy and
their power of accreditation, it will be the death of knowledge.

Received on Tue Aug 25 1998 - 19:17:43 BST

This archive was generated by hypermail 2.3.0 : Fri Dec 10 2010 - 19:45:27 GMT