The Shakespeare Conference: SHK 12.0254  Friday, 2 February 2001

From:           William Proctor Williams <This email address is being protected from spambots. You need JavaScript enabled to view it.>
Date:           Thursday, 01 Feb 2001 19:16:05 -0500
Subject:        Early English Books Online

Last autumn I gave it as my opinion that Early English Books Online
(EEBO) was not a very valuable research tool.  As a result of those
comments Bell & Howell offered me a free, no- strings-attached trial use
of the EEBO database.  I would like to stress that my somewhat revised
view of that database has not been influenced by any sort of inducement,
financial or otherwise.

There is no doubt that EEBO delivers digitized images of STC/Wing books
of high quality and with a relatively high standard of accuracy.  When,
in a few years' time, all of the images have been digitized, it will
deliver an online version of the images now found on the Early English
Books microfilm (EEB) in most cases superior to what one might read and
print in the microforms departments of most research libraries.  For
institutions which never subscribed to the EEB project, or who did so on
an intermittent basis, this may be a way to make all these works
available to their readers at a lesser cost than buying the microfilms,
even if they could do that.  The stated cost of this was $93,000, but in
January 2001 that price was removed and the following statement replaced
it: "Costs for the Bell&Howell EEBO image product depend on whether or
not institutions already own the Early English Books microfilm; for
almost all institutions, the price falls at less than a dollar a
volume.  At 125,000 volumes, this is a very large corpus so it seems
fair to view the price in that context."  (see:
http://www.lib.umich.edu/libhome/eebo/licensing.html).  Whether or not
this makes the cost to an institution more or less than the original
$93,000 I am unable to say.

Bell & Howell, in conjunction with the University of Michigan Libraries
and the Bodleian Library, has a linked project (EEBO-TCP) which plans to
provide full-text versions of about 26,000 works selected from the
approximately 150,000 items in STC/Wing.  Subscribing to this full-text
database costs $50,000 spread over five years (see:
Bell & Howell believes that the full-text and image databases are
complementary and so an institution is very strongly encouraged to
participate in both projects and thus the total bill would probably be
in the neighborhood of about $140,000.  However, Bell & Howell informs
me that the pricing is more flexible than this and can be/is based not
only on how much of EEB the institution has already purchased, but the
FTE of the institution (one assumes this excludes non-educational
institutions such as independent research libraries) and other factors
as well.

Of course one gets at the images through searching and such searching
relies on the cataloging of the EEBO project.  It is here that one
encounters some rather serious problems, in part because EEBO appears to
rely heavily on the cataloging of another electronic database, the
English Short-Title Catalog (ESTC), but does so in anomalous ways.  For
example, format statements are never provided though they appear in
ESTC.  In my testing of the database I have found a number of
difficulties but I will not bother the membership with them though I
will be happy to elaborate for those who contact me by private email.  I
will only say here that fairly simple searches have funny results. If
one searches for "Ling" in imprints in the advanced search mode or
"Ling" in the simple search "Keyword in any field (e.g, philosophy,
MacBe?)" option one gets 47 items the first time and 81 the second time,
but what one never gets is either Q1 or Q2 _Hamlet_ though Ling appears
in the imprint of both as publisher (a search of ESTC for Ling in the
imprint yields 61 items, but among them are Q1 and Q2 _Hamlet_). This
unsatisfactory aspect of the EEBO database means that until the
cataloging is improved, and I must add that Bell & Howell has indicated
that they are working on improving it, using EEBO for teaching or
research will be very much like using a blunt instrument.  It also means
that just as ESTC cannot really be safely used without concurrent use of
the printed catalogues and indices for the STC/Wing period, so EEBO
cannot be safely used without concurrent use of them and probably of
ESTC as well.  For example, EEBO, like ESTC, omits all references to
copyright entry and copyright transfer to be found in STC.  Since there
is no index for the Stationers' Register for the period before 1640 STC
has always partially fulfilled that role (there is a complete index to
the Stationers' Register for the period from 1640 to 1709 so the
omission of this information by Wing and by ESTC is less of a problem).
In any case, if the scholar wishes to trace the copyright history of a
work from this period she/he will receive no help from either EEBO or
ESTC [I should like to note that when I raised the question of the
omission of copyright information with both ESTC and EEBO I was told by
both that I was the first person ever to have asked the question].  In
fairness I should say that ESTC has vastly expanded our identification
of surviving copies of pre-1700 books.  Of course, EEBO only identifies
the copy being filmed/digitized.  However, a virtue EEBO and ESTC have
is the ability to do subject and keyword searches.  Although my
experience tells me that no one should regard such searches as complete
(i.e., a search for a term or phrase will not always, or perhaps ever,
produce all the items containing that term or phrase), it is the sort of
searching of books printed before 1700 which would be laborious, perhaps
impossible, using only the printed databases.

Finally, A. W. Pollard said, when the STC was first published in 1926,
"it is a dangerous work for anyone to handle lazily."  What was true in
1926 is doubly true in the brave new electronic world of 2001.  It would
be very unwise for any scholar to treat these databases (ESTC and EEBO)
as definitive.  The printed ones, and particularly the RSTC, have the
virtue of decades of concentrated revision and also the considerable
virtue that most of the work has been "book-in- hand" bibliography.
Alas, the conversion of all this work into electronic form is treading
dangerously near the corruption of more than a century's careful work in
the field of Early English books.  The danger here is increased as fewer
and fewer graduate programs in the humanities require, or even offer,
courses in physical bibliography.  As more and more teachers and
students enter the stream with less and less knowledge about how early
books are constructed, the deficiencies of these seductive online
databases cannot, I fear, do much to advance knowledge and may lead to
actual error.  I hope that constructive criticism, such as I intend this
to be, will overcome market forces and institutional loyalties and that
those in charge of both ESTC and EEBO will draw back from the brink,
draw breath, and resolve to revise and correct what they have already
done in their massive tasks, to correct what they know to be errors and
problems, and to accept corrections from those who use their databases.
The field is one which has always benefited from collegial scholarship
and one which has suffered from commercialization.

A very much longer version of all this will appear as an article in
_Analytical & Enumerative Bibliography,_ New Series 12.1 sometime in the
next few months.

William Proctor Williams

Subscribe to Our Feeds


Make a Gift to SHAKSPER

Consider making a gift to support SHAKSPER.