Catalogs, Card—and Other Anachronisms

by Karen Coyle

Preprint. Published in: The Journal of Academic Librarianship , Volume 31, Issue 1 , January 2005, Pages 60-62, doi:10.1016/j.acalib.2004.12.001


"It is fortunate for those who have the use of a library if their number is so small and their character so high that they can be admitted to the shelves and select their books on actual examination. As that is often not the case, a catalogue becomes necessary, and, even when it is the case, if the books are numerous there must be some sort of guide to insure the quick finding of any particular book. The librarian can furnish some assistance, but his memory, upon which he can rely for books in general use, is of no avail for those which are sometimes wanted very much, although not wanted often." [1]

The library card catalog was one of the great inventions of the 19th century. Today we see it as old-fashioned, and there are active users of libraries today who never used a card catalog, but just as today's users get satisfaction out of typing in a keyword and getting results from a search engine, card catalog users approached the catalog as a box of treasure in which previously unknown things could be found. Those of us who did cut our bibliographic teeth in the days of oak-drawered cabinets and buff-colored cards didn't question them; their format had fixed long before our time and was so familiar that it seemed almost a state of nature.

In fact, the card both contained and enforced some interesting "technologies," if you use the term in a broad sense to mean things of human invention. The catalog card is both artificial and abstract: if you would hold up a library card side-by-side with the book it represents, you would be struck by how un-alike they are. As a representation of the book, the card has little that the general reader would see as familiar. Not only is the pleasing design of the cover totally absent, few of the elements that you find on the card are to be found on the book, at least not in the form in which they are represented on the card.

A Biography of Melvil Dewey



Wayne A. Wiegand




American Library Association

Chicago and London, 1996

  Wiegand, Wayne A., 1946
     Irrepressible reformer : a biography of Melvil
  Dewey / Wayne A. Wiegand.
  Chicago, American Library Association, 1996
A Book... ... and its Card

When you contemplate the sheer artifice of the card, it's a wonder that library users have managed to adopt the library view of the bibliographic universe, and often without any formal training. And although we have progressed beyond the card catalog to online catalogs, we are still working within some of the constraints of that technology. For the sake of our users it would be a good idea to decide to leave the card behind us, once and for all.

Some History

Most libraries in the U.S. had their main catalog in book form when the American Library Association, urged on by Melvil Dewey set the standard for the size of cards in 1877 [2], The widespread use of cards did not begin until over 20 years later when the Library of Congress began printing cards for sale to libraries. Cards allowed the virtual duplication of the book so that it could be sought from multiple points of view, and the printed cards meant that at each entry point the user would see a full citation.

The card catalog was necessarily linear, and linearity in most libraries was equivalent to "alphabetical order."[3] The points by which cards in alphabetical catalogs were ordered were headings, and those headings needed to be printed exactly as they would be filed in the catalog and searched on, from left to right. With the card, the heading had to serve both as the sort key and the display form. This brought us the comma-delimited personal name, i.e. "Smith, John F." and strange subject headings like "History, Ancient, in literature" and "Mary, Blessed Virgin, Saint." It also brought us the concept of main entry, essentially one author and one title. Main entry implies that there will be added entries, which are entered at the bottom of the card like so many distant cousins.

The technology of the card dominated the library world for over one hundred years. It went unquestioned until about 1980 when the combination of the existence of the Machine Readable Cataloging (MARC) format, originally designed to facilitate on-demand printing of cards, and increased computing power made it possible to consider putting the card catalog data into a database where it could be searched from a keyboard and displayed on a screen. Unfortunately, these new computerized catalogs were being created with exactly the same data that was used to create the card catalog, with its emphasis on headings and on alphabetical order.

The Online Catalog

Here's an entry in a not atypical online library catalog. It has eschewed the card format for a labeled format, but the difference between this display and a card display is not as great at it may seem.

Author : 	Durant, Will, 1885-
Title : 	The age of reason begins : a history of European civilization in the period of Shakespeare, Bacon, Montaigne, Rembrandt, Galileo, and Descartes: 1558-1648 / by Will and Ariel Durant
Published :New York : Simon and Schuster, 1961.
Description : 	xviii, 729 p., [20] leaves of plates ; 25 cm.
Series : 	Story of civilization : pt. 7
Notes : 	Includes index. Maps on lining papers.
Notes : 	Bibliography : p. 649-659.
Subject : 	Thirty Years' War, 1618-1648.
Subject : 	Europe -- History -- 1517-1648.
Authors : 	Durant, Ariel.

We still have a main entry at the top of most of our displays, and secondary entries or added entries below. So although Will Durant and Ariel Durant shared the title page on their Story of Civilization series beginning with volume VII [4], in library catalogs Ariel Durant gets second place, relegated to the bottom of the card in card catalogs or at the screen in online catalogs. The two authors are brought together in the statement of responsibility (that many of us know as the $c subfield of the 245), but that statement is less prominent than the headings are. Yet the statement: "by Will and Ariel Durant" is both clear and precise. It is also what is on the book, both the cover and the title page, therefore it is what the reader who has the book in hand sees.

Another leftover from the card is that we are still displaying the headings in an inverted order, as if that form determines how they will be searched. In online catalogs, the display form and the indexing and sorting forms of a heading are entirely separate, although the latter two are derived from the former. In fact, both sort keys and indexes as they are stored in a database are significantly different from the display heading. Typically, although it can vary by system, headings and their sort and index forms can look like this:

DisplaySort keyKeyword index terms
Doyle, Arthur Conan, Sir, 1859-1930doyle arthur conan sir 1859doyle arthur conan sir
Tolkien, J. R. R. (John Ronald Reuel), 1892-1973tolkien j r r 1892tolkien j r r john Ronald reuel
Transportation -- Accidentstransportation accidentstransportation accidents
Transportation accidentstransportation accidentstransportation accidents

What this means is that we have already separated the display and retrieval functions. So why are we clinging to the form of heading that was required when display and retrieval had to be one and the same? The answer is in our cataloging rules, which still require catalogers to create the card entry, and in the fact that all of our millions of records and dozens of different system designs we are heavily invested in this form of catalog record. It will take a willful act of amnesia for us to contemplate the possibilities for a library catalog as if the last 200 years of librarianship had not taken place.

Rethinking Bibliographic Display

If we can suggest that we would like to move beyond the card metaphor that has locked us into some now outdated practices, what would the replacement be? What would our catalogs look like? First, we should think about what we want our catalogs to be, and how we can best serve the user. We may wish to have the flexibility to present the user with a display that reads:

The Age of Reason Begins: a History of European Civilization
in the Period of Shakespeare, Bacon, Montaigne, Rembrant, Galileo,
and Descartes: 1558-1648. By Will and Ariel Durant, 1961.


Will Durant; Ariel Durant
   The Age of Reason Begins: a History of European Civilization
in the Period of Shakespeare, Bacon, Montaigne, Rembrant, Galileo,
and Descartes: 1558-1648. New York, Simon and Schuster, 1961. 

At the same time, if we are producing an alphabetical listing, both names should appear under "Durant" not "Will" or "Ariel." This is all actually very simple because today we can code the heading in such a way that all of these functions are equally possible using the same data. If we have a name coded as:

     family name: Durant
     forename(s): Will

we can create sort keys with the family name preceding the forenames, and we can create indexes that allow a search on "durant will" or "will durant". This is because the functioning of the catalog and the display form of names, titles or other information can be separated in ways that they were not for the card catalog. So we are now free to create displays that are independent from our indexes and from any need to create ordered displays. One advantage, however, of the card was that the display that the user saw was the same in all library catalogs. As we break away from the rigid card catalog display, we do need to consider that users are served by a degree of uniformity in their bibliographic world. Should there be a standard display, or a set of standard displays? Should the library display take on any of the promotional aspects of the online bookstore, such as cover art or marketing blurbs?

Retrieved Sets

In the card catalog, users encountered cards in order, one at a time. There was no retrieved set in that catalog. There was no time when a group of cards would be gathered from disparate locations in the catalog based on a word or a group of words unrelated to their original order. The card catalog was always about headings; in the online catalog, searches always result in a retrieved set of records that then must be displayed to the user. The question is: in what order should those items be displayed? What is a useful order when a library user has done a keyword search, or even a search on a heading? Most library online catalogs have only two possible orders for retrieved sets: by date, either ascending or descending; and by main entry. Whenever items are put in an order, the order is only helpful when it is somehow related to the user's information need. So chronological order with the most recent publications first may be very helpful in some disciplines, but probably not for the user seeking 19th century literature or the works of Shakespeare, which have been republished throughout the centuries. And with a subject search that retrieves hundreds of items, it's doubtful that the presentation of those works in main entry order has any meaning to the user doing the search. Some union catalogs are experimenting with ordering works by "popularity," that is by the items that are most held. This may produce a granular ordering for a portion of a retrieved set, but does nothing for all of the items that are held by only one library. This is a failing of many ranking systems; there simply is not enough information to rank an entire set. Ranking of the type that is used in Internet search engines is at least in part based on the frequency of terms in full text works. Applying that ordering to bibliographic records has already proven to be less useful. Of course, some day we may have the full text of all of the books in digital form, like Amazon's "Look inside the book" feature, and could use that information in our catalogs. Since library records have classification numbers, we could provide retrieved sets in a classified order, or grouped by general subject area. Which order we would choose could depend on the nature of the catalog or even the nature of the retrieved set, with small sets being presented in a single order and large sets getting broken into groups by topic, date range, or some other order. The main thing is that the order needs to be relevant to the query, and that's what we are not providing in our catalogs today.

What about headings? Are they entirely a thing of the past, or will we still have a need to have headings and to present them to users in order? If so, what should that order be? Would anyone want to rank authors by the number of works they have written? Library catalogs often have a somewhat disembodied heading browse that displays the headings alphabetically apart from the bibliographic items where they are found. Is this useful? If headings are to be presented in alphabetical order, it would be interesting to learn which of the displays below are favored by catalog users:

Inverted formNatural form
Asimov, Isaac
Bradbury, Ray
Cadigan, Pat
Vinge, Joan
Vinge, Verner
Willis, Connie
Isaac Asimov
Ray Bradbury
Pat Cadigan
Joan Vinge
Verner Vinge
Connie Willis


If we do manage to free ourselves from the card we could find ourselves participating in a true revolution in our field. Just as the card catalog allowed the library to provide user service far above what had previously been possible, a catalog based on the technology of computing and freed from the constraints of the card could revitalize the library for current and future generations. This is not a pipe dream; it is only a matter of opening our minds to the technological tools that are before us. Surely there are some explorers in our midst willing to blaze a new trail


[1] Cutter, C. A. Library Catalogues. In: Department of the Interior, Bureau of Education. Public libraries of the United States of America, their history, condition and management. Washington, Government Printing Office, 1876. p. 526.

[2] Wiegand, Wayne A., Irrepressible reformer : a biography of Melvil Dewey. Chicago, American Library Association, 1996, p. 54

[3] There was some experimentation with catalogs in classified order, but these proved difficult for the public to use and often required alphabetical indexes to make them accessible. see: Cutter, C. A., op cit., pp. 529-532. The rise of alphabetical or dictionary catalogs coincided with the introduction of the Dewey Decimal Classification which created a classified order for the books on the shelves, supported by an alphabetical index to the classification for public access. see: Dewey, Melvil. A Classification And Subject Index, For Cataloguing And Arranging The Books And Pamphlets Of A Library, Amherst, Mass, 1876.

[4] "Mrs. Durant's part in these concluding volumes has been so substantial that our names must be united on the title page." Durant, Will and Ariel Durant. The Age of Reason Begins. New York, Simon and Schuster, 1961. p. viii.

The copyright in this article is NOT held by the author. For copyright-related permissions, contact Elsevier Inc.