8

Suppose you have two libraries that use the same classification system and have not merged their respective databases. Could it be possible that the exact same book could be assigned different call numbers if taken from one library and added to the other?

I am asking this question because I am drawing up a home-library database application, and the client would like the ability to assign call numbers (even multiple ones). I have not been able to find an answer online, and I hope that I do not have to travel to the next state over to find a library not connected to the ones around I live.

I imagine the book would most likely be assigned to the same top-tier category and most likely the same second subcategory in the respective system, but, for example, the Library of Congress system incorporates the author's name and a quasi-decimal number which fine-tunes the book's location. And it seems without any external reference the assignment of this part of the call number is made on the fly for any particular library.

So, I have a couple questions. First, the title's sake. Second, is the way I'm thinking about this correct? And if I'm wrong, why?

8

Normally, when already available, Library of Congress classifications are usually found after the title page, and should thus be constant across different libraries. For publishers.

In any case, this should be a fairly easy hypothesis to test (at least to a certain degree). Most university library catalogues are on-line and freely available. You could sample a cross-section of titles from different subjects and see what happens: my suspicion after a few cursory checks of my own is that call numbers do vary somewhat, but primarily at the "secondary level" (in identifying the author, rather than the subject).

  • One would expect LC numbers to be invariant. However... an uncatalogued book with a title something like "Classical Problems in Mongolism" arrived at a certain university library, and it was given a number appropriate to Classical Mongolian. The library declined to correct the error, holding that "at least it is catalogued". – user6726 Jun 12 '15 at 15:55
5

A call number is meant to be a unique identifier for a particular book (or other work); that is, a particular physical object, typically found on a shelf somewhere. These identifiers are often structured, containing encoded information about the subject, author, publication date, etc. If a library has multiple copies of the same text, then they will typically have different call numbers - say, via a suffix "(copy 1)", "(copy 2)", etc. They are different physical objects. They are tracked separately: one copy was purchased, and another was donated; one is on loan and another is having its spine repaired; and so forth.

It is devoutly to be wished that different libraries, following the same standard, will assign their identifiers in the same way. Reality being what it is, we can't guarantee that this will always happen, even if it is the common case. However, your question is about something different: when a book is moved from one library to another, what happens to the call number? There are two things going on here:

  1. Assigning a new number. Since you have a new physical object, it needs an identifying number. The multiple copy example above shows one instance where the least-significant part of the call number would have to change. Ideally, other changes would not be needed in order to make the number conform to the new library's system. So the question is: how much do you trust the old library to get it right? If the answer is anything less than 100% then there should be some way to assign a new number.
  2. Keeping track of the old number. Even after renumbering, it might be useful to record the book's former call number. Provenance information is generally felt to be a Good Thing to have - certainly, in the case of items with any value, or which are externally indexed. (For example, bibliographies of the works contained in a particular library will often include the shelfmarks.) This may not be a concern for your database, or it might not have to be part of the main database system - perhaps "how I obtained this book" information could be retained elsewhere.

If you are dealing with MARC, then you can look up how it handles coding of former locations, multiple call numbers, and so on, as there are standard ways of recording this information.

  • (1/2) "a unique identifier for a particular book" - I'd disagree with this. There are libraries that assign unique-per-item call numbers, but this is rare; most commonly it's a classification number, and it's usually seen as acceptable to have a dozen different titles all classified at, say, 551.352 SMI - the reader can easily see which one they want now you've got them to this location. Suffixes help to provide order but they aren't required to be unique. – Andrew Jun 13 '15 at 19:17
  • (2/2) For unique-item tracking you'd usually assign an item number, which is usually stored as a barcode inside the cover (or an RFID tag) rather than a human-readable call number on the cover/spine. These will be unique, hopefully within the entire system not just the one site, and provide tracking ability on a per-item basis for recording acquisition, repairs, etc. – Andrew Jun 13 '15 at 19:18
4

A bit late to answer the OPs question, but for the record... the answer is "usually they will be the same, but there is no requirement and there are often good reasons to vary".

Factors which may cause a different number to be used by different libraries:

  1. Different classification systems. Usually, different branches of a given library will use the same classification system, but this is not guaranteed, especially when different institutions have merged.

  2. Different versions of the classification system. Classification systems are frequently revised, and not all libraries update their existing shelfmarks when an updated edition is issued. This paper discusses the example of the "eugenics" subject, which has been assigned to 18(!) different locations in Dewey at various times.

  3. Different local focus. If individual libraries have a specialised focus, they may choose to classify books in a different way. For example, I used to deal with a small departmental library focused on classical studies. A book on, say, the military of the Roman Empire would be shelved under 937 as a particular subset of ancient history in the main library, but the second copy in the specialist library would be shelved under 355 as a book on military history - this was the most convenient approach for that particular site.

  4. Different local practice. There are a number of aspects of the classification system that are effectively optional. For example, a smaller library may decide to use shortened numbers rather than the longer, more "correct", ones. "Cutter codes" (the quasi-decimal bit based on author name) are often assigned in an idiosyncratic way, and some libraries will use a local or arbitrary system for these. Fiction/literature also causes problems, as using the "pure" classification system for this is relatively rare outside large libraries; many will use a much simpler system, and it is very likely this will be inconsistent between individual libraries.

2

I took @aeismail suggestion literally and tried to find an older geodesy textbook. C.F. Baeschlin's Lehrbuch der Geodäsie that was published in 1948. I checked MELVYL, the University of California's system, which is now connected with WorldCat. They listed all copies as QB283.B2 for the Libary of Congress (LC) system, but had other identifiers including an OCLC number. I then checked Ohio State's library catalog which listed their copies as QB283.B33. Both systems do have the same OCLC number.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.