© 2009 James P.

Ascher

Progressing Toward Bibliography; or:

Organic Growth in the Bibliographic
Record

Much of traditional book cataloging is influenced by a touch-it-once mental-

ity, where work is completed in as few steps as possible. By avoiding unnecessary
revision, libraries can process materials quickly and benefit from economies of
scale. Valuable staff time is preserved to process additional materials. However, the
touch-it-once mentality is problematic when dealing with special collections ma-
terials, since they often differ between manifestations and are primary sources for
research that informs their description. This paper discusses the idea of “progres-
sive bibliography,” or proceeding from minimal to fuller descriptions, as an intel-
lectually valid and pragmatically essential methodology. It examines some already
existing approaches and discusses a few of the challenges and further areas of
research. While the idea of positive accumulation of knowledge is old, computer-
ized tools and modern information theories enables us to streamline this process to
the benefit of patrons and scholars, and so managers of these collections have new
tools to tackle ever-increasing backlogs of underprocessed materials.

The issue of achieving access to special collections materials has received a great
deal of attention in recent years. The concepts of rare book cataloging, however,
are often firmly seated in traditions that predate the computer. The traditional
ways of cataloging and documenting have not typically included the possibility
of enhancing the description as knowledge changes over time. Yet, in many cases,
not everything is known about a document before it is cataloged, especially since
undercataloged materials have not been previously available to scholars. Modern
scholarship can put the mundane into a new light that highlights its importance.
Thus, the work of the rare book cataloger is often the beginning of more substan-
tial bibliographical research, and there is a desperate need for new ways to inte-
grate computerized tools into the construction of scholarly databases. As knowl-
edge progresses, it needs to be included in our progressing descriptions, which in
turn help progress scholarly knowledge.

95
96 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

Earlier Literature
Progressive description can be adapted to various economic environments to do
rare material cataloging inexpensively while still doing full description as needed.
Scholarly databases, such as the catalogs maintained by libraries, are a public
good. However, libraries face increasingly difficult financial situations. The eco-
nomic realities that most cataloging departments face include cut or transferred
cataloging positions, so having catalogers work on detailed descriptions for each
item as it is acquired becomes less feasible for many institutions. M. Winslow
Lundy suggests using noncatalogers, who are subject experts and bibliographically
sophisticated, to provide initial access to collections.1 Lundy’s procedure uses the
supervision of catalogers and is based in part on the idea of “Phased Cataloging”
as described by Lawrence J. McCrank.

McCrank uses survey methodology to describe and analyze the rare book cata-
loging workflows of various institutions. The survey supports a hierarchy of ten
descriptive levels: directory description, collection description, short-title cata-
loging, item level standard cataloging, standard rare book cataloging, advanced
variant descriptive cataloging, full-text retrieval for edited materials, facsimile
retrieval for unedited materials, document retrieval in actuality, and retrieval at the
lower levels plus contextual information.2 These levels represent progression from
library-focused directory information and short-title cataloging to the descriptions
created by textual critics and bibliographers. In McCrank’s observations, optimal
processing levels are generally assigned as goals to parts of collections. Although he
doesn’t discuss a method, he suggests that description and surrogation are ongo-
ing processes and that “revision is always taking place.” Interestingly, McCrank
observes that the cost per item seems to tie most strongly to the size of the collec-
tion and the geographic location. The per-item cost becomes slightly cheaper as
collections reach ten to twenty-five thousand items and then begins to grow with
collection sizes. McCrank hypothesizes that this relates to the quality of cataloging
done at larger institutions.

In archival circles, the work of Greene and Meissner continues to make an

impact on both the economics and the efficiency of descriptive work.3 They
suggest that some access to everything should come before detailed access to
anything. In archives, that idea translates to using original order, if present, and

1. M. Winslow Lundy, “Providing Access to Uncataloged Special Collections with In-Process Re-
cords,” Cataloging & Classification Quarterly 45, no. 1 (2007): 39–58.
2. Lawrence J. McCrank, “The Bibliographic Control of Rare Books: Phased Cataloging, Descrip-
tive Standards, and Costs,” Cataloging & Classification Quarterly 5, no. 1 (1984): 27–51.
3. Mark Greene and Dennis Meissner, “More Product, Less Process: Pragmatically Revamping Tra-
ditional Processing Approaches to Deal with Late 20th-Century Collections,” The American Archivist 68.2
(fall/winter, 2005): 208–63.
 Progressing Toward Bibliography 97

accepting brief access until more detailed access is warranted. Archival descrip-
tion is built with an awareness of progression as part of the history of the
methods; but progressive description has not been fully implemented at most
institutions, nor has progression been adapted to rare books and other special
collections materials.

The Association of Research Libraries identified developing a continuum of de-

scription as an important step toward providing intellectual access to collections for
scholarly purposes.4 The recommendation suggests: “View the description of con-
tent as a continuum, take advantage of existing standards that can address the wide
varieties of materials and needs, and promote flexibility in selecting the standard to
be applied to any materials.” This desire for an ability to work on description along
a continuum of description is echoed in the work of the Library of Congress in
their Report on the Future of Bibliographic Control.5

The Report, recently released by the Working Group at the Library of Congress,
focuses strongly on the need to enhance access to rare, unique, and other special
hidden materials. Access to hidden materials is an urgent topic for libraries as cus-
todians of our nation’s cultural heritage and as we move into an increasingly digital
realm, where physical manifestations become more removed from the physical
location of scholarship. Part of the statement is particularly germane to the work
of rare book catalogers:

2.1.2.1 All: Adopt as a guiding principle that some level of access must be
provided to all materials as a first step to comprehensive access, as appropri-
ate. Allow for different cataloging levels depending on the types of docu-
ments, their nature, and richness.

Although this might seem revolutionary considering the prevailing touch-it-

once mentality for rare book cataloging, it actually has a historical basis. In 1908,
Falconer Madan, a librarian at the Bodleian Library, almost certainly faced with a

4. Association of Research Libraries, Exposing Hidden Collections Conference Summary, (Sept. 8–9,
2003), available online at http://old.arl.org/collect/spcoll/ehc/EHC_conference_summary.html [ac-
cessed 31 March 2009]; Mark Dimunation, Exposing Hidden Collections Conference Summary: Recommenda-
tions from Monday’s Breakout Sessions (2003), available online at http://old.arl.org/collect/spcoll/ehc/
Dimunation_Summary.html [accessed 31 March 2009]; Association of Research Libraries, ARL Special
Collections Task Force Report (Sept. 13, 2002), available online at http://old.arl.org/collect/spcoll/tforce/
status.html [accessed 31 March 2009]; Barbara M. Jones, Hidden Collections Scholarly Barriers: Creating Ac-
cess to Unprocessed Special Collections Materials in North America’s Research Libraries (2003), available online
at http://old.arl.org/collect/spcoll/ehc/HiddenCollsWhitePaperJun6.pdf [accessed 31 March 2009].
5. Library of Congress Working Group on the Future of Bibliographic Control, On the Record: Re-
port of the Library of Congress Task Force on the Future of Bibliographic Control (Washington, D.C.: Library
of Congress, 2008); Library of Congress Working Group on the Future of Bibliographic Control, Report
on the Future of Bibliographic Control: Draft for Public Comment (Washington, D.C.: Library of Congress,
Nov. 30, 2007).
98 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

surfeit of uncataloged books and a dearth of time, proposed to address his problem
with Degressive Bibliography.6 Restated in 1923:

We have learnt not to regard books in a library as all equal in importance

and all to be treated alike, as if they were a rank of drilled soldiers. The
lesser books must stand back, and the greater be brought into prominence.
We must make use of mental perspective, and provide Degressive Descrip-
tion… The idea is that different periods of printing and different classes of
books should meet with correspondingly varying treatment.7

Madan splits time into Early (to 1550), Middle (to 1800), and Modern (after 1800),
and classes books as Important, Interesting, and Ordinary. Claiming this solution is
important for any library holding more than one thousand volumes, he then goes
on to give a scheme, and example descriptions, that catalogers can use along with
their judgment to ensure the correct level of description.

The Library of Congress’s principle of “cataloging levels depending on the types

of documents, their nature, and richness” is a clear reformulation of Madan’s
Degressive Description. The revolutionary reinterpretation is supported by the
technology now available. Madan, and his successors, were working with codex
catalogs and, later on, card catalogs. To revise these descriptions, physicality
became a major obstacle. Pages had to be erased and leaves had to be added, or
cards must be replaced. Since every book would receive description based on its
age and importance eventually, it was more efficient to handle a book once and
describe it for all time. This is a valid time efficiency if knowledge is canoni-
cal and well-defined. That is, if early books would always be early, important
authors and printers would always be the same and no new important actors
would be identified. Thus, the cataloger would know these things about the
book at the time of cataloging and set them in stone.

This is similar to the model espoused by Bennett and other catalogers from the
print tradition.8 Bennett states “…since rare books will attract inquiry and must
sooner or later be properly described, it would appear to be a question of covering
an initial expense or absorbing a running expense as it accrues.” Furthermore, “It
has been suggested to me that I enlarge upon this, showing that the initial expense
is less than the running expense of temporary cataloguing and eventual revision.”

6. Falconer Madan, “Degressive Bibliography,” Transactions of the Bibliographical Society, IX (Oct.

1906–Mar. 1908): 53–65.
7. Falconer Madan, “Standard Descriptions of Printed Books,” Proceedings and Papers, Oxford
Bibliographical Society 1 (1923): 55–64.
8. Josiah Q. Bennett, The Cataloguing Requirements of the Book Division of a Rare Book Library, Occa-
sional Papers, no. 3 (Kent, Ohio: Kent State University Libraries, 1969).
 Progressing Toward Bibliography 99

In the current cataloging environment, the cost savings of the touch-it-once men-
tality has proved effective in the cooperative cataloging environment and provides
the financial impetus for the standard records of the Program for Cooperative Cat-
aloging. In this environment, one participant describes something correctly once
and then other institutions can use the same record. This saves time and money,
since books will not have to be cataloged more than once. Yet, touch-it-once cata-
loging can be problematic, since many of the characteristics of rare materials are
unique to the holding repository, and the awareness of certain characteristics may
not even exist when the initial cataloging is done.

For example, consider the case of information about stereotype foundries often
found on the verso of the title page in nineteenth-century books. This information
indicates the shop that built the printing surfaces and is thus an important piece of
industrialized printing history. Very reasonably, this sort of information is excluded
from the catalog record by the Library of Congress Rule Interpretations of the
Anglo-American Cataloging Rules,9 which are not primarily concerned with rare
books or bibliographical research into the nineteenth century. This makes sense in
a cooperative cataloging environment where libraries are only concerned with the
content of the book, but it omits characteristics of primary sources in industrial
print history. Unfortunately, the records are structured with a touch-it-once mental-
ity so that updating this information in a shared record might be inaccurate for
books which have a different stereotyper, but share the record. This harkens back to
the days of purchasing printed cards where revision, and addition of information,
was done only locally.

Our technological environment has, however, changed. Information is fungible

with the ease of a mouse click and keyboard stroke; catalog records can be auto-
matically updated; databases can be linked; and knowledge can be intertwingled
and transcluded, ideas originated by Nelson.10 Nelson argues that computerized
text can be repeated and digitally connected between various systems. Transclud-
ing is the process of repeating information from one context in another, essen-
tially providing a two-way footnote to another document. Intertwingularity is the
awareness that knowledge can be connected by cross-connections, through linking,
transclusion, and other such citations. These ideas have proved extremely powerful
and useful already. Nelson’s early work led to the development of hypertext, which

9. Cataloger’s Desktop, Library of Congress Rule Interpretations (Washington, D.C.: Library of

Congress), available online at http://desktop.loc.gov/ [accessed 27 August 2009]; Cataloger’s Desktop,
Anglo-American Cataloging Rules (Chicago: American Library Association, 2005), available online at
http://desktop.loc.gov/ [accessed 27 August 2009].
10. Theodor Holm Nelson, “Xanalogical Structure, Needed Now More Than Ever: Parallel Docu-
ments, Deep Links to Content, Deep Versioning, and Deep Re-use,” ACM Computing Surveys 31.4es
(1999): 33; Theodor Holm Nelson, “The Heart of Connection: Hypermedia Unified by Transclusion,”
Communications of the ACM 38 (1995): 31–33.
100 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

formed the basis of Tim Berners-Lee’s work on HyperText Markup Language, that
has become the lingua franca of the Internet. In library cataloging, evidence of
transclusion can be found in authority records that provide a linking mechanism
for materials based on a standard title and that can then connect different databases.
The intertwingled catalog would thus be a network of bidirectional connections
allowing users to explore a bibliographic galaxy. The technology for this exists
and has been implemented for various projects such as Wikipedia, WorldCat, and
ZigZag.11

Furthermore, our intellectual environment is dynamic. Knowledge is not canoni-

cal; we have multiple perspectives: postmodernism, regional studies, Kuhnian
paradigms, print culture, reader response, the sociology of the text, and many other
ontological and epistemological theories. Heaney points out that the role of catalogs
is much larger than the original purpose of inventory control: “The problems arise
from the fact that library catalogs do not function solely as descriptive lists of books:
they function also as elements in library management systems and, in the electronic
age, as sophisticated information retrieval tools.”12 Sensitivity to changes over time is
the core of the progressive methodology, and something that library cataloging has
overlooked for years.13 While current staffing and knowledge might lead to minimal
description, the description generated by a rare book cataloger begins the process
that can lead to a critical edition, progressing along a bibliographical continuum.

The revolutionary part of the idea of progression as traced through this literature
is that we finally have the technology to efficiently “provide [access] to all materi-
als as a first step to comprehensive access,”14 and this is a good idea because of
the changing face of scholarship. In other words, degressive description becomes
progressive description, which becomes more detailed as the importance of the
work grows. It also reacts to changes in scholarly knowledge, as D.W. Krummel has
suggested.15The cataloger is no longer the judge, jury, and executioner of the re-
cords of the material of the rare book room, but a gatekeeper and ally to the whole
Gutenberg galaxy of information, ideas,and connections.

11. Theodor Holm Nelson, “ZigZag (Tech Briefing),” HYPERTEXT ’01: Proceedings of the Twelfth
ACM Conference on Hypertext and Hypermedia. Århus, Denmark (New York: Association for Computing
Machinery. Available online at http://portal.acm.org/toc.cfm?id=504216&coll=portal&dl=ACM&typ
e=proceeding&idx=SERIES399&part=series&WantType=Proceedings&title=Conference%20on%20
Hypertext%20and%20Hypermedia. [Accessed 26 August 2009].
12. Michael Heaney, “Object-oriented Cataloguing,” Information Technology and Libraries 14.3 (1995):
135–53.
13. Michael Heaney, Time Is of the Essence (Aug. 11, 1997). Available online at www.bodley.ox.ac.uk/
users/mh/time978a.htm. [Accessed 20 March 2009].
14. Library of Congress Working Group on the Future of Bibliographic Control, On the Record: Re-
port of the Library of Congress Task Force on the Future of Bibliographic Control (Washington, D.C.: Library
of Congress, 2008), 32.
15. D.W. Krummel, “On Degressive Music Bibliography,” Notes 2nd Series, 56.4 ( June 2000): 867–78.
 Progressing Toward Bibliography 101

Bibliographic(al) Information
Rare books are often of value because they can be located in a particular biblio-
graphical milieu. Changes between printings and editions can indicate the social
reception of the text. Offensive text might be canceled, lazy compositors may
add significant errors, and the physical resources used by the printing house may
change. Furthermore, lists of materials can provide evidence clarifying many
questions, such as “Which books were printed at the same house at the same
time? What types of books were being sold?” These questions can be addressed
in a number of ways including interrogating the object directly through physi-
cal bibliographical research. Physical bibliography is interested in the physical
state of the book and what this implies about the conditions of its creation, and a
highly detailed physical bibliography is generally given the honor of being called
a descriptive bibliography.16 Of course, the facts of the physical nature of a book
are endless—lines can be individually measured, broken type can be counted,
bindings can be identified, and such ad nauseam. To put a limit on the amount
of data collected in a physical bibliography, Professor Bowers suggests that one
should focus on a specific purpose and scope: “It should have a unified subject, a
definite purpose expressed in its arrangement and in its treatment of the books
described so that a shaping intelligence guides the work.”17 This differs from the
goal of cataloging that must describe materials that exist within a particular col-
lection to fit within a constellation of different databases that adhere to various
standards. Libraries don’t necessarily have a unified subject, definite purpose, or a
complete authoritative collection in all areas, so to attempt a library bibliography
would be foolhardy. Even in cases where a library has a comprehensive collection,
the records usually adhere to external standards such as those for OCLC WorldCat
or the Anglo-American Cataloging Rules. Thus, the apparatus of a catalog differs
from a bibliography.

Bibliographical scholarship can thus be distinguished from cataloging in both scope

and methodology. While a catalog describes the holdings of a particular institution,
bibliographical scholarship attempts to formulate some sort of understanding of
the deeper nature of the work or text. The activities are, however, closely connect-
ed. Tanselle describes the connection:

The two are naturally related pursuits, and the interests of all who are con-
cerned with books are best served by a spirit of cooperation between them;
the split which threatens to make them continually more incompatible does
no one any good … Bibliographers and cataloguers, and many other people

16. G. Thomas Tanselle, “Descriptive Bibliography and Library Cataloguing,” Studies in Bibliography
30 (1977): 2–57.
17. Fredson Bowers, Principles of Bibliographical Description (New York: Russell & Russell, 1962), 18.
102 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

as well, constantly consult both catalogues and bibliographies; the two

kinds of works are necessarily different, having different aims, but they are
both parts of a larger undertaking—the recording of intellectual products
and their physical embodiments.18

This connection was also emphasized by John Alden, who suggested “bibliographi-
cal cataloging” as describing an object with an awareness of the study of bibliog-
raphy.19 He argued that special cataloging provided the opportunity for scholars to
find significant patterns by looking at whole collections and using the descriptions
of librarians as starting points. Josiah Q. Bennett argues that the cataloger is re-
sponsible for putting a rare book within its “bibliographical milieu”20 by describing
connections and relationships between it, published bibliographies, other catalogs,
and other versions of the work that are likely to exist.

In both these cases of bibliographical awareness in cataloging, two things are

happening: one for current scholarship and one for future bibliographical scholar-
ship. For current scholarship, the cataloger must serve due diligence to existing
knowledge about the material in question, since users of the materials would
presumably be interested in this aspect. This aligns with the identify objective
described by Svenonius, “to confirm that the entity described in a record corre-
sponds to the entity sought or to distinguish between two or more entities with
similar characteristics.”21 Thus, the additional detail in bibliographical cataloging
is motivated by users of rare books and special collections materials, who have an
exceptional awareness of different characteristics of entities.

For future bibliographical scholarship, later scholars can use the library’s
description as a starting point for further developing some sort of scholarly
resource. This method is described in many textbooks, including that by D.C.
Greetham,22 and is important because a sizable amount of material exists within
libraries as handled by librarians. Through familiarity and close observation,
catalogers are often aware of details in which future bibliographical scholars
will be deeply interested. The practice of cataloging gives the experience sought
in bibliographical laboratories like McKerrow’s and conforms to William A.
Jackson’s suggestion of handling many books to become familiar with what is

18. Tanselle, “Descriptive Bibliography and Library Cataloguing,” 2.

19. John E. Alden, “Cataloging and Classification,” Rare Book Collections, ACRL Monograph Number 27
(Chicago: American Library Association, 1965), 65–74.
20. Bennett, Cataloguing Requirements, 6.
21. Elaine Svenonius, The Intellectual Foundation of Information Organization (Cambridge, Mass.: MIT
Press, 2000), 20.
22. D.C. Greetham, “Finding the Text: Enumerative and Systematic Bibliography,” in Textual Scholar-
ship: An Introduction (New York: Garland Publishing), 1994.
 Progressing Toward Bibliography 103

normal.23 Bibliographical information that develops through handling the mate-

rial can be recorded in library catalog records that are shared widely and thus
contribute to the accumulation of knowledge on these subjects.

The bibliographical information recorded in catalog records is, however, not static.
Lagoze points out that “objects may change over time, morphing into new versions
of their former selves or into things altogether different.”24 In rare books, this fluid-
ity is certainly true, but the challenge lies in keeping the description from becoming
too complicated.25 Books may become disbound through repeated handling or be
rebound in new forms after initial cataloging. Colors may fade and sewing might
weaken to reveal bibliographical evidence that was obscured by tight bindings.
Additionally, different curators and catalogers might possess expertise in different
issues of intellectual and cultural history and so be able to contribute knowledge
where previous librarians were mute. Once the tenant of recording bibliographical
information is accepted, revision and change over time becomes necessary since
bibliographical scholarship is not static.

Time awareness in revisions is visibly absent from early cataloging and bibliograph-
ical methodologies. This makes sense when the description is published in a book
and is difficult to revise; but, as Fattahi points out, the computerized bibliographic
environment dramatically changes the sort of work that can be done.26 With the
advent of computerized bibliographic databases, the work of libraries has become
more involved and cooperative, while at the same time the amount of information
needing cataloging has grown rapidly. This growth has obscured the advantages of
computer-based organization of rare books with additional work in more volumi-
nous areas. However, the fungibility of computer-based information provides an
amazing opportunity to develop better work flows that can adapt to scholarship.
Nicolas Barker eloquently puts it:

23. Steven Escar Smith, “‘A Clear and Lively Comprehension’: The History and Influence of the
Bibliographical Laboratory,” in Teaching Bibliography, Textual Criticism and Book History (London: Picker-
ing & Chatto, 2006), 33–38; William A. Jackson, “The Education of a Bibliographer,” in Records of a Bibli-
ographer: Selected Papers of William Alexander Jackson (Cambridge, Mass.: Belknap Press, 1967), 47–50.
24. Carl Lagoze, “Business Unusual: How ‘Event-awareness’ May Breathe Life into the Catalog,”
Conference on Bibliographic Control in the New Millennium (Washington, D.C.: Library of Congress, Oct.
19, 2000). Available online at http://lcweb.loc.gov/catdir/bibcontrol/lagoze_paper.html. [Accessed 20
March 2009].
25. Carl Lagoze, “Keeping Dublin Core Simple: Cross-Domain Discovery or Resource Description?”
D-Lib Magazine 7.1 (2001). Available online at www.dlib.org/dlib/january01/lagoze/01lagoze.html. [Ac-
cessed 20 March 2009].
26. R. Fattahi, The Relevance of Cataloguing Principles to the Online Environment: an Historical and Analyti-
cal Study: A Thesis Submitted in Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the
School for Information, Library and Archive Studies, University of South Wales (Sidney, Australia: University
of South Wales). Available online at http://profsite.um.ac.ir/~fattahi/thesis1.htm. [Accessed 6 March
2009].
104 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

… [the computer] can tell you where to find books, what is in them, and
much else, that books have been forced, against their nature (try reading
the back of today’s title-pages), to tell us themselves. Let the computer do
this: what then? The books can be in random order. If order is needed, let
it be, like Samuel Pepys’s, order of height; it looks well and is better for the
books.27

Ultimately, the fungibility and reusability of computer-based information allows

progressive description to more accurately reflect the nature of scholarship and the
issue of time in cataloging.28

The Fungibility of Knowledge

Knowledge and scholarship have always been mutable. Thomas Kuhn famously
describes the social construction of paradigms in scientific revolutions,29 while Ian
Hacking explores how the idea that knowledge is socially constructed has permeat-
ed modern discourse and that writing papers about social construction has become
a sort of game.30 Bella Weinberg argues that the depth and nature of all catalog-
ing is relative to the particular collection in which something belongs.31 Norman
Anderson likewise points out that cataloging can be in need of correction since it
might be biased toward language, romanization, religion, type of library, descrip-
tion versus subject, form of heading, and fiscal concerns.32 A correct canon implies
the existence of other canons, and modern scholarship has become increasingly
interested in finding these alternative canons. In particular, bibliographical scholar-
ship is highly revisionist and frequently explores previously unnoticed aspects of
the book. The intellectual paradigms change with each succeeding generation of
scholars, who must find new things to talk about.

An easy example of an intellectual change is the awareness of nineteenth-century

publisher’s cloth book designers. Sue Allen pioneered research into this area, which
has been expanded upon by many art history and book history enthusiasts.33 The

27. Nicolas Barker, “Libraries and the Mind of Man,” in A Potencie of Life: Books in Society, the Clark
Lectures 1986–1987 (London: British Library, 1993), 193.
28. However, the limits of this fungibility have interesting implications for the study of the history
of cataloging. See Matthew G. Kirshenbaum, Mechanisms: New Media and the Forensic Imagination (Cam-
bridge, Mass.: MIT Press, 2008).
29. Thomas Kuhn, The Structure of Scientific Revolutions (Chicago: University of Chicago Press, 1962).
30. Ian Hacking, The Social Construction of What? (Cambridge, Mass.: Harvard University Press, 1999).
31. M. Bella Weinberg, “A Theory of Relativity for Catalogers,” in Cataloging Heresy: Challenging the
Standard Bibliographic Product (Medford, N.J.: Learned Information Inc., 1991), 7–11.
32. Norman Anderson, “The Non-neutrality of Descriptive Cataloging,” in Cataloging Heresy: Chal-
lenging the Standard Bibliographic Product (Medford, N.J.: Learned Information Inc., 1991), 15–28.
33. Sue Allen and Charles Gullans, Decorated Cloth in America: Publishers’ Bindings, 1840–1910 (Los
Angeles: UCLA, Center for 17th- and 18th-Century Studies, William Andrews Clark Memorial Library),
1994.
 Progressing Toward Bibliography 105

work of nineteenth-century designer John Feely is thus easy to identify. In most

libraries, samples of his characteristic gold-stamped bindings can be found on some
popular mass-market books. Likewise, one can find other books designed by him
but rebound in buckram. When the volume was rebound in buckram, the libraries
probably did not consider the graphical work of Feely significant. Since the book
was from the nineteenth century, a comparatively recent era, and about a popular
topic, his designs were probably looked upon like much of the cover art of newer
books: possibly interesting, but not essential to scholarship. Thus recent, unimport-
ant books received minimal treatment, which we would expect from the standard
degressive bibliographical point of view.

However, since the books were first cataloged and received minimal treatment,
people have studied the process of the creation of these bindings and the artist
has been identified. The images of John Feely have been used to explore the social
reaction to various books and ideas, and his aesthetic is studied as a serious decora-
tive art. Many catalog records, even for rare books, don’t mention him as the cover
designer, largely because no one knew who he was (or cared) when the original
records were created. Now institutions cherish their publishers’ bindings and often
construct special digital libraries to show off their holdings. Publishers’ bindings
have now become a worthy topic of study; and, thus, catalog descriptions of books
ought to progress to support this research. Since the descriptions already exist, the
only way to proceed is through progression. This can be done by adding informa-
tion to a catalog record, but approaches to progression other than continually
updating catalogs also exist.

Approaches to Progression
An example of an essentially revisable descriptive source is the English Short Title
Catalog.34 It was begun as the Short Title Catalog of Redgrave and Pollard35 and
continued with the unifying purpose of comprehensively identifying every separate
printing, issuance, state, and edition of every book printed in the English language;
description is thus limited to a short one and the sufficient details needed to distin-
guish the various states of books. Other information is rightly omitted because it
does not address the purpose of this bibliography.

As a cooperative resource and as a product of human hands, errors creep into the
description. Stephen Tabor in his 2007 article sampled 150 ESTC records and found
an average rate of errors (that would hinder identification) of 17 percent.36 Among

34. English Short Title Catalog. Available online at http://estc.bl.uk/. [Accessed 1 September 2009].
35. Alfred W. Pollard, G.R. Redgrave, William A. Jackson, F.S. Ferguson, and Katharine F. Pantzer, A
Short-Title Catalogue of Books Printed in England, Scotland, & Ireland and of English Books Printed Abroad,
1475–1640 (London: Bibliographical Society, 1976).
36. Stephen Tabor, “ESTC and the Bibliographical Community,” The Library 8 (2007): 367–86.
106 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

the oldest records, all books printed up to 1640, he found a surprising 52 percent of
errors in the records. An immutable, describe-it-once-and-shelve-it mentality would
have no way of correcting these errors. The errors would quickly be regarded as
truth since they are in such an esteemed bibliography. Bibliographically identical
items would be identified as something new since their description did not match
the errors present in the old description. Tabor concludes that we must be diligent
in revising records and continually reviewing books as we come across changes.
The strength of our digital fungible data is that we have the eyes of decades of
contributors—librarians, scholars, both, and amateurs—to ferret out the mistakes
that come along.

Not every revision needs to be due to an error. Scholarly information sources can
provide information that is useful for adding to records. Exhibition catalogs can
provide a wealth of information about materials that are held by a library and the
institutional focus on a particular topic so that the information can be discovered.
Yet the information in the exhibit catalog may not make it into the bibliographic
catalog. For example, an exhibit catalog on machine-era books might bring out
specific facts about the books but not contain full catalog entries, and the method
for describing the materials may not be compatible with traditional cataloging.
Consider a copy of Sermons on Various Subjects by Joseph Lathrop, printed by Isiah
Thomas on both handmade woven paper and laid paper. This is unusual, since both
types of paper were used together for only a decade or so. However, the fact was
apparently unrecorded until the publication of an exhibit catalog. Certainly this
would be useful information for a historian of American printing or paper, but they
would have no access to it since it was only publicized in a small print run exhibit
catalog.

Every time someone publishes a bibliography, designs an exhibition catalog, makes

an online exhibit, or adds an entry to a database, there is more information avail-
able about a book. Though the bibliographies may have a specific scope, the facts
they give about the works are still true for the works. One cannot just arbitrarily
add these facts to national level catalog records, however, because a large number
of entities use the records and abide by various rules and guidelines. One approach,
similar to attaching social tags to the catalog records, is to develop links to these
other sources with other rules and guidelines, except that the librarians decide
whether to ingest the data. This idea of the librarian collecting this supporting data
predates the prevalence of the computer.

In the time before hyptextual information and the ability to intertwingle text,
Josiah Bennett called the supporting data a “bibliographical file.”37 A file would be

37. Bennett, Cataloguing Requirements, 39.

 Progressing Toward Bibliography 107

kept on each book, and clippings, notes, data, photocopies, or anything else related
to the book would go in the file. The file itself served as a research resource for the
book, and in time the catalog record might be revised based on the contents. These
resources would certainly be provided to scholars requesting more information
about a particular book. However, scholars would have to inquire about a specific
book and certainly could not automatically search the data in these files, and the
information would only be whatever the librarian had found.

In the age of networked and harvestable data, we can share far more information
than Bennett could. Scholars, enthusiasts, and librarians routinely generate useful
digital data about books held in trust by special collections. Katz recommends that
libraries use scholars to expose collections and materials to a wider audience,38
while Schreyer encourages the use of students in enhancing description.39 Tabor
argues that some problems in distinguishing rare books might be better handled
by scholars than catalogers. In-house knowledge can be recorded in an accessible
form, the catalogs from exhibitions can be encoded so that their research is avail-
able, local and visiting scholars can provide their notes for encoding, courses taught
in the manner of Rare Book School, through direct interaction with materials, can
accumulate data about books and check them repeatedly as exercises, scholarly
publications can be encoded, and many other activities can generate information.

Once information is recorded in digitally accessible repositories, it can be linked

to the catalog records for library materials. Readers of the digitally accessible
information could trace the link to the record for the actual book, which could
be used to locate the material for use. This approach would be much like the
one-way scholarship of years past, only more efficient—a reader of one of these
resources could always find the materials in a collection by using the citation.
Digital linking would only add to the ease of paper citations. More data can be
accessed quickly by linking catalog records back to the entries in the bibliog-
raphies and exhibit catalogs. Then, a scholar researching a particular book in a
collection might be alerted that some preliminary investigation had already oc-
curred, without having to travel to the collection, and at the same time could see
what is unique.

This approach becomes significant when multiple sources of information exist.

In such a case, a researcher can find a reference in one document (for example,
about the paper used in a book), then trace the type of paper on the record for the
particular item, and then trace the item to another reference (about, for example,

38. Exposing Hidden Collections Conference Summary, 2003.

39. Alice Schreyer, “University of Chicago Explores Library-Faculty Partnerships in Uncovering Hid-
den Collections,” ARL 251 (Apr. 2007): 4–8.
108 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

the binding). This linking would truly allow the exploration of the sociology of
the text, as D.F. McKenzie advanced, because a scholar could trace themes from
one textual object to another in the socially related objects.40 Furthermore, a single
catalog record, representing a physical item, would contain a large amount of in-
formation. The institution may have started with one preliminary record; but, after
doing a few exhibits, instruction sessions, inviting researchers, and other activities,
one will suddenly have a great deal more information about our artifact. One can
then revise our catalog record within the national finding aids that utilize it with
local information.

Other approaches to progression have been attempted in catalogs. The master and
institutional record concept recently brought to OCLC from RLG allowed institu-
tions to enhance records with special information and to share that information.41
This allows institutions to progress their description without damaging the basic
description, which is more useful for copy cataloging. However, the system does
not provide a method for progression, since it has not been built with rare book
cataloging in mind; it simply provides a mechanism that could be used for progres-
sion.

Another system for progression has been provided in the work of the Rare Books
and Manuscript Section of the Association of College and Research Libraries
(RBMS) in the development of Descriptive Cataloging of Rare Materials (Books), or
DCRM(B).42 The standard is part of a series of descriptive cataloging standards for
rare materials. This descriptive standard builds on the shared work of the Anglo-
American Cataloging Rules, which is the standard cataloging rules used by many insti-
tutions. By building on AACR2, DCRM(B) provides an additional set of descriptive
rules that are used to make enhanced descriptions of special collections materials.
Thus, the step from AACR2 to DCRM(B) is a method of progressive description,
based on the principle of remaining compatible with previous work.

Another important tool maintained by RBMS is a series of thesauri.43 The Sec-

tion maintains six different flavors: binding, genre, paper, printing and publishing
evidence, provenance, and type evidence. These controlled vocabularies contain

40. D.F. McKenzie, “Bibliography and the Sociology of Texts,” in The Panizzi Lectures 1985 (London:
British Library, 1986).
41. Liz Bishoff and Glenn Patton, “Master Bibliographic Record Vs. Local Bibliographic Record—
Who Needs What?: An OCLC Perspective,” in Cataloging Heresy: Challenging the Standard Bibliographic
Product (Medford, N.J.: Learned Information Inc., 1991), 85–90.
42. Association of College and Research Libraries, and Library of Congress, Descriptive Cataloging of
Rare Materials (Books) (Washington, D.C.: Cataloging Distribution Service, Library of Congress, 2007).
43. Association of College and Research Libraries. RBMS Controlled Vocabularies: Controlled Vocabular-
ies for Use in Rare Book and Special Collections Cataloging (Chicago: Rare Books and Manuscripts Sections,
Association of College and Research Libraries, 2005). Available online at www.rbms.info/committees/
bibliographic_standards/controlled_vocabularies/index.shtml. [Accessed 27 August 2009].
 Progressing Toward Bibliography 109

narrowly defined terms that can be used to systematically describe collections. The
terms are created based on the needs of librarians and maintained as a semanti-
cally valid ANSI standard thesaurus. These terms can go beyond exposing hid-
den collections and instead expose hidden aspects of collections. By naming and
exposing facts that are esoteric to the subject of the material, catalogers can enable
researchers to find materials for which they would otherwise need stack access and
browsing time.

While catalogers can apply these facts to their records to further enhance their
work, noncatalogers with certain areas of expertise can also be trained to add
thesaurus terms to catalog records. Institutions can select a small set of terms so
that people inexperienced with rare materials can be trained to identify certain
characteristics. The work then becomes answering yes or no to a series of discrete
questions rather than the difficult work of placing an item within the vast biblio-
graphical milieu. The smaller size of the set of terms would make this work easier.
It would still essentially be the work of a librarian, but simplified.

Another set of familiar tools are those that form what is generally termed Web 2.0,
or the social Internet. These tools are social tagging, user reviews, and comments,
which have been enabled in many next-generation catalogs. OCLC has begun
to provide some of this functionality in their open WorldCat database that now
includes comments and reviews. These tools have not been explicitly applied to the
idea of progressive bibliography, but they can be used to accumulate data about li-
brary materials. While these tools show great promise, there is a great deal of work
to be done before they can be integrated into library work flows with both the clar-
ity and authority that rare book cataloging generally requires. Rolla compares user
tagging and Library of Congress Subject Headings (LCSH) and finds that, while
tagging provides useful access, the LCSH provides types of access not covered by
these tags.44 Rolla continues to discuss how tagging could be improved with ideas
from the LCSH and the LCSH could be improved with ideas from tagging. It is
not surprising that LCSH provides better access in certain ways since professional
librarians build it, using their knowledge of information architecture, while users
build tag sets, using their knowledge of what they want.

While tagging and subject cataloging address the fourth level of McCrank’s phased
description (item level standard cataloging), they do not help with the higher levels
of observing variants and building toward textual criticism. For these levels of
description, tools like blogs can provide an information infrastructure. As described
by Wikipedia—which is generally enlightening regarding new computer technology

44. Peter J. Rolla, “User Tags Versus Subject Headings: Can User-Supplied Data Improve Subject Ac-
cess to Library Collections?” Library Resources & Technical Services 53, no. 3 ( July 2009): 174–83.
110 RBM: A Journal of Rare Books, Manuscripts, and Cultural Heritage

(but not necessarily other things)—“A blog … is a type of website, usually main-
tained by an individual with regular entries of commentary, descriptions of events,
or other material such as graphics or video.”45 Blogs, in a library context, could allow
the continuous accumulation of data about particular topics. A number of library
curators are already using tools like blogs to highlight collections, but the blogs have
no link back to a centralized information system. Adopting track-back functionality
(an implementation of two-way links) in a library catalog could provide an easy way
for this information to be accumulated in a digital bibliographic file.

Conclusion: Progressive Bibliography, or Organic Growth in the

Bibliographic Record
Improving the quality of rare book cataloging through progressive description
promises to improve research for many scholars. Scholars using the work of
the catalog as a beginning to progress to new bibliographical description clearly
benefit, but recently scholars of rare books and book history have also begun do-
ing research using the data within library catalogs. Norbert Schürer uses library
circulation records, based on cataloging records, to study the construction of
knowledge,46 while Lishi Kwasitsu uses the National Union Catalog of Australia to
explore the history of publishing in Victoria.47 These sorts of studies can be greatly
improved by enhancing the raw data that the studies use, and progressive descrip-
tion is a method to continually improve information contained in catalogs.

To enable the sort of organic growth suggested here, much work is needed.
Paradigms for information access are firmly entrenched, and fully conceptualized
methods to allow the enhancement of records as proposed above do not yet exist.
Likewise, the question of authority and trust—whom to allow to enhance re-
cords—is one that needs to be explored by further research. Should libraries allow
any use to link data to official bibliographic records? Or would this open them up
to manipulation by commercial advertisers and political groups seeking to manipu-
late the catalog? If libraries want to allow linking to official records, but control the
process, what would the workflow be? Experimentation and research into these
issues could provide us with ways to incorporate progressive description into bib-
liographic work. Ultimately, the computer has given us the ability to enhance the
structure and content of records, but our workflows have yet to adapt. The idea of
progressive description is the first step toward better systems that should function
in our twenty-first–century information environment.

45. “Blog,” Wikipedia. Available online at http://en.wikipedia.org/w/index.php?title=Blog&old

id=310845748. [Accessed 30 August 2009].
46. Norbert Schürer, “Four Catalogues of the Lowndes Circulating Library, 1755–66,” The Papers of
the Bibliographical Society of America, 101.3 (Sept. 2007): 329–57.
47. Lishi Kwasitsu, “Publishing in Victoria, Australia, 1851–1900,” The Papers of the Bibliographical
Society of America 101.2 (June 2007): 205–19.