A drawer of cards pulled out in a card catalog.

Card Tricks: The Decline & Fall of a Bibliographic Tool

By Stephen J. Greenberg ~

View of the Card Catalog in the NLM Rotunda from the second floor balcony.
Aerial View of the NLM Rotunda and Card Catalog, ca. 1964
National Library of Medicine #101445862

There was a time, not so very long ago, when card catalogues were pretty much synonymous with libraries. You really could not imagine one without the other. As late as the 1960s, some library buildings were architecturally designed around their physical catalogues. The current main building of the National Library of Medicine is a good example.

The move away from such catalogues began a long time ago. Henriette Davidson Avram, working with a team at the Library of Congress in the late 60s, developed the MARC format that would pave the way for all future computerized library catalogues. It has now been over thirty years since NLM removed its main card catalogue, although some remnants persist. After all, no major research library ever had just one card catalogue; there were always several, all for different purposes. Cards were the physical manifestation of the collective librarian memory, and there were a LOT of things to remember.

The retrospective conversion of most card catalogues to electronic formats, and the overall passage of time, have made it increasingly necessary to describe to younger visitors what a card catalogue is (or rather, was). There are, however, some of us still in the profession who remember cards, if not always fondly.  One needed specialized training to type cards.  It was an art in itself, and it has become a lost one. Particularized use of standard punctuation marks, initials, and seemingly bizarre use of spacing were all part of the cataloguers’ codebook. Nothing was ever random, however. The seemingly idiosyncratic spacing and punctuation allowed the cataloguer, and the informed researcher, to identify the author and title on a given card, even if the language and the character set were strange to the user. Going back even further, when cards were handwritten, special penmanship had to be learned, a throwback (even then) to the late Middle Ages, when each royal government office had a prescribed script.  One could identify the source of the document simply by recognizing the script style (for example, the so-called “Chancery Hand”).

There are tales to be told that are rapidly fading to folklore.  Even casual users of old catalogues are aware that the cards don’t fall out of the drawers, even if the drawer is inverted, although flipping the drawers is a cheap thrill when giving a tour to the uninitiated. The reason is simple.  Each card has a hole neatly punched through its center, at the bottom of the card.  A long brass rod, running the depth of the card drawer, goes through the hole and secures the cards in place. The rod is removed to add a new card, and that is where the folklore comes in.  Novice cataloguers were once directed to put their new cards in the proper place in the drawer, but NOT to remove the rod until their work could be checked by more senior staff.  This was called “cataloguing above the rod.” When the newbie could be trusted to get it right without supervision, there were permitted to “drop” their cards right in: “cataloguing below the rod.” Who will remember THAT in fifty years?

On purely bibliographic grounds, it is hard to be nostalgic for card catalogues. For simple searching in an unfamiliar field, they cannot compete with a modern online catalogue. In a traditional card catalogue, a simple book would be represented on several cards: an author card (usually the “Main Entry,” of which more in a minute), a title card, and subject cards, maybe three to six per book. If one knew the proper terms for your topic (the “controlled vocabulary”), one could usually get a good start on any research project.  If one didn’t know the precise terminology, there might be “see’ or “see also” notes to guide one along. But sometimes, the subjects could be maddeningly elusive. Variant ways of spelling or even word arrangement could lead to frustration, and not a few jokes funny only to librarians.  One old chestnut is the proper way to index the Holy See (also called the See or Diocese of Rome, which is the jurisdiction of the bishop of Rome, AKA the pope). Does the noun come first, or the adjective? Traditional cataloguing leads to the lovely note, directing the researcher to the proper heading: “See, Holy. see Holy See.”

People sit at computer terminals on tables next to the card catalogs and bookshelves.
Online Catalog Terminals Alongside the Card Catalog, ca. 1985
National Library of Medicine #101445872

Cards have the problem of a very limited number of searchable fields. One cannot search a card catalogue to locate all of the Paris imprints from 1745 (for example).  But assuming one can search the appropriate fields in the computer record, it’s easy to do that search with an online catalogue. And therein lies the rub of any comparison between cards and computers: cards are only searchable using the strategies already put in place when the cards were typed.  No new search strategy can be created, nor can multiple searches be combined.  There’s no Boolean searching with cards.  With computer searching, such elegant strategies are easily applied, and these days taken for granted. As computers grow ever more powerful, their ability to handle full and free text searches renders any other option almost laughable.

When the Anglo-American Cataloging Rules (AACR) were revised in the late 1970s, computers were forcing serious changes in the librarian mindset, but they were not the ubiquitous force they are today. A 1970s VAX mainframe had less computing power than a modern smartphone. The public hearings about the proposed AACR2 at American Library Association meetings led to some odd moments as the card culture collided with the computer horizon. For example, for many, many years, the “Main Entry” card in the catalogue (the one card that combined ALL of the information known about the item) was the author card. Traditionally, that card was the author’s real name: Samuel Clemens, not Mark Twain; Mary Ann Evans, not George Eliot; Eric Blair, not George Orwell; Karen Blixen, not Isak Dinesen, et. al.  AACR2 wanted to change that, and put the “known” name as main entry. There are stories of voices raised and tears shed; the amount of work required for that card catalogue revision would have been boggling, albeit a boon to the less experienced, researcher. But with modern online catalogues, it really doesn’t matter. ANY search will take you to the full record.

Interior view of the reading room of the Army Medical Library.
The Library’s reading room in the “Old Red Brick” ca. 1910
National Library of Medicine #101415763

There are occasional news reports about card catalogues being rescued on the eve of their destruction. We are told that there is valuable information that will be lost if the cards are discarded.  Perhaps that is true, although that speaks to the care with which the cards were “re-conned” (retrospectively converted) in the first place. Some of NLM’s last remaining cards contain shelving information in the “Old Red Brick” building that NLM left in 1962, and which has since been demolished to make space for the Hirschhorn Gallery. One can never be too careful; once they are gone, they are truly gone. And that is why every remaining NLM card is examined before its fate is decided.

Black and white Photograph of a white man.Stephen J. Greenberg, PhD, is Head of Rare Books and Early Manuscripts for the History of Medicine Division at the National Library of Medicine.



  1. An enjoyable read Stephen. Thanks for sharing your knowledge with those of us who only remember the card cabinets as young patrons. Ours is still in the library and it continues to be an obsolete but comforting fixture. That said, someone just walked by to look up something in the big, print dictionary. Go and figure!

  2. Getting rid of reliable “old” technology! “Be careful what you wish-for.”
    or is it that, “that for which you wish?”

    1. I’m with Winston Churchill on not ending sentences with prepositions: “This is the type of errant pedantry up with which I will not put.”

  3. Great reminder of the limits and costs of technologies we were once nostalgic about, especially as we struggled with the transition. I still struggle with finding cross references in computer subject searching but once I find the right term the computer is greater resource. Well done.

  4. Steve, this was lovely. We are soon to be rid of our last card catalogs here at Mizzou, and while there is a fondness for them as bastions of library-ness (and they are pretty furniture) the fact is that the few we have remaining for shelflisting purposes haven’t been used or updated for years. So they are going. We are going to invite everyone to “drop a drawer” of cards in the recycle bin and have a big cake to “celebrate.” Not sorry the days of cataloging over the bar are gone, necessarily, but I will miss my stalwart friends.

  5. I go back before typed catalog cards — although I must have typed 100,000! When I began my first job at the main Detroit Public Library in 1962, I would sometimes find cards handwritten in beautiful Palmer calligraphy in the card catalog. Thanks for the memories. My career ended in 2001 with no card catalogs in sight.

    1. I could not write proper Palmer if my life depended upon. We could have a whole other thread about penmanship, and how cursive writing is (and isn’t) being taught today.

  6. I began my career in libraries at 15 years old as a page who shelved books and filed cards. As a professional, I revised the filing of clerks and filed the shelf list cards. I do not for one moment miss any of it!

    I do need to correct one claim, however. “But with modern online catalogues, it really doesn’t matter. ANY search will take you to the full record.” Only if someone has entered the information in the first place, either as a secondary entry in the bibliographic record or (preferably) as a cross-reference in the authority file that is linked to the bibliographic file. In other words, it is just as accurate to say that “catalogs are only searchable using the strategies already put in place when the bibliographic files were created.” Some human being — some librarian — has to analyze the item and create the records and establish the links among the files.

    1. Agreed: a bad record is a bad record, whether it’s a card or online. Moreover, some of the shelving information created during the Old Red Brick building days is still in use today, which is why we have to check the cards against the online catalogue, to make sure that we are not discarding that the only remaining record of that information. But with a properly done online record (done by a real live trained librarian), there are a far greater possibilities for searching.

  7. As a reference librarian, I do not miss trying to teach students how to find things in the card catalog, especially not the rule about words starting with Mc being filed as if the were spelled Mac. 😀

  8. A great article. Thanks. When I was (as the only librarian) converting the catalog to digital format, I had the print shop pad the cards to use as note cards. I couldn’t bring myself to just toss them. I hated to give up the physical cards. You’re correct that you can do great searches using the digital catalog but most of my users weren’t even able to use the cards, much less be interested in using the computer. And, yes, I’m talking about teachers .

  9. Loved reading this on the eve (literally) of my retirement from over 40 years of teaching cataloguing in Melbourne, Australia. Thankyou. I am going to miss it

  10. When I came to my current library as Director – 36 years ago! – I realized that the cards had been filed completely wrong, using a locally grown system. It was easier to automate than to refile the whole card catalog!

  11. Also enjoyed this walk down card catalog lane! As Suzanne above… I was a page in a public library in 1969 (50 years ago!) and learned the art of filing cards above the rod… and then taught the same to library students when I became a director. When we gave up the card catalog in my present hospital library, I tried to find uses for the beautiful oak card catalog… one was filing cassette tapes in them. But oops… those are now history too! Thanks for a great article. 🙂

  12. I learned how to type the cards properly in grad school and what a bear they were. By the time I got to grad school I had been a library clerk for years and looked forward to the day I could file below the rod. Imagine my sorrow when I got to the Carnegie Library of Pittsburgh and NO ONE got to file below the rod. Also, anyone who misses the card catalog should have to spend the day putting the drawers back together after some enterprising person figured out how to pull the rods and dump the drawers. I do remember a story from one of my professors about the rioting at Columbia in the 60s–police were sent to guard the card catalogs from the rioters. It’s nice to remember the days when those gorgeous pieces of furniture meant that much. Thanks for a great article.

  13. Technology marches on. One sees web sale listings for old card catalogue cabinets at thousands of dollars.

  14. Yes, electronic catalogs are richer and more accurate, but some information is lost when card catalogs disappear. The simple image of the entry, before you even read it, indicates if it is an older book or not (hand-writen card, ink or ball-point pen, which kind of typewriter). The additions can be full of information. And there is also a certain kind of serendipity that is lost in the computer. Librarians ought to think twice before simply destroying the old card catalogs.

    1. NLM carefully reviewed its cards before the big recon thirty-plus years ago, and some cards were kept. The current review is also a question of many eyes and hands, and some cards will remain. Serendipity (of a different kind) still exists with an online catalogue – – – one can browse a subject heading while limiting by date, for example. The skills change with the technology.

      1. Thank you for your answer. Yes, skills change with the technology, but they don’t simply replace older skills. They are an additional layer of knowledge. I’m in favour of keeping the old card catalogs, especially in libraries with historical heritage. They are an archive (I am an archivist and historian…but not nostalgic). Of course, serendipity is also present in online catalogs, but the discoveries are different.

  15. I wonder where I can buy a good used card catalog…such an amazing and comforting piece of history 😊.

  16. I earned good money at UT Austin’s Perry Castaneda Library filing cards into the card catalog. I came upon the most interesting titles while filing and once my shift was done I would go up into the stacks to take a look at those books. The only thing I miss about the card catalog is that serendipity factor as one thumbed through cards – I rarely come upon something unexpected and curious while using our OPAC.

  17. Thanks for this. It was a nice visit to my past as a cataloguer (1980 to 2016). I arrived at the end of the card era and the coming of AACR2, and left soon after RDA arrived.

  18. A thought-provoking reflection on the decline of card catalogs as a bibliographic tool. This article delves into the evolution of information management, shedding light on the transformative journey of libraries.

  19. Fascinating read! The evolution and eventual decline of card catalogs as a bibliographic tool is a testament to the ever-changing landscape of information management.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.