How do old books get ISBN numbers?

[John]

I have the 1958 Pantheon edition of Dr. Zhivago. I can find no ISBN number on this book, yet Amazon lists a 1958 Pantheon edition with the ISBN 0002611503. How is this possible? I thought ISBN numbers were first issued in the mid-1960s. Is it possible that the edition I have is different from the edition Amazon has?

[rivka, Librarian Moderator]

Amazon has a lot of errors. A Google search for that ISBN indicates that it actually corresponds to the 3rd edition of the 1958 translation, published in 1966. I have altered the GR record accordingly.

Unless this edition is close enough, you'll have to enter your edition by hand.

[John]

Thanks Rivka.

(1) Regarding ISBN numbers, then, is it correct to say that any book with an ISBN has been published no earlier than the mid-1960s?

(2) What kind of Google search do you do that reveals the correct information on a book with an ISBN number? I went to http://www.google.com/ig , put the ISBN number (0002611503) in the box next to "iGoogle" and hit the Google Search button. The first hit took me to Amazon.co.uk where I'm told that this book was published by Harvill Press in Dec 1957. The second hit took me to Amazon.com which tells me that this book was published by Pantheon in 1957 (this was the original info on GR that you corrected). The third hit takes me to a library site in Newcastle, UK which claims this book is a 3rd edition published by Collins in 1966. The fourth hit takes me to tomfolio which gives the publication date as 1957 and doesn't mention a publisher.

BTW, your comment "Amazon has a lot of errors." is the understatement of the year.

[Seizure Romero]

Excuse my jumping in here...I am aware that my name isn't Rivka.

1) Not necessarily. Reprints of older books will be assigned ISBNs by the publisher. That is why books in the public domain (like most 'classics': Shakespeare, Chaucer, etc.) are issued by different publishers and have different ISBNs. Conversely, modern small presses issue books without ISBNs quite often (think poetry chapbooks).

2) Unless you have the book in hand or are looking at a scan of the copyright page, there really isn't a way to guarantee the accuracy of your information. The one thing you must remember is that Amazon's sole purpose is to sell books. The accuracy of their information is directly connected to that. Pre-internet, used book dealers actually had to know something about their wares. Now, any chucklehead with a computer and a garage full of books can call him/herself a book dealer. Many of these people are selling through Amazon and other online book sites; unfortunately, just like here on GR, they are able to input information no matter how inaccurate. If the book in question is not currently in print, Amazon isn't selling it directly and therefore has no vested interest in verifying or correcting any of the information attached to that record.

A few links which may be of interest:

http://www.isbn.org/standards/home/is...

http://en.wikipedia.org/wiki/ISBN

[rivka, Librarian Moderator]

What kind of Google search do you do that reveals the correct information on a book with an ISBN number?

One where I assume that the most logical and/or consistent result makes sense. ;) Occam's Razor and all that.


Reprints of older books will be assigned ISBNs by the publisher.

But said reprints are new editions, published post mid-60s. Which is exactly what John was asking. Since the ISBNs was invented in 1966, I don't know how it would be possible for any book published before then to have one! Unless the time-travelers have been messing with the timeline AGAIN! ;)

[This Is Not The Michael You're Looking For]

But said reprints are new editions, published post mid-60s. Which is exactly what John was asking. Since the ISBNs was invented in 1966, I don't know how it would be possible for any book published before then to have one!

I'm not entirely sure this is correct (speculation follows). While it is certainly true that anything physically published prior to 1966 cannot have an ISBN, many books are reprinted...the reprint is not considered an edition change, but later reprints likely would have been given ISBNs.

For example, I have a copy of Eight Stories from the Rest of the Robots by Isaac Asimov. This book was originally published in 1964. The edition which I have (Pyramid Books) was first published in January 1966 (according to its title page), thus predating the use of ISBN (which was developed in '66 and implemented in '67). However, the specific book which I have is from the 7th printing (May 1974) and it does have an ISBN. Thus, that same ISBN would be used to refer to all of the printings of this book, even though some of them were done before the ISBN existed...it is considered the same book. A new edition is a different type of thing than a new printing.

Random book trivia: How can you determine the printing of a recent book? Some books (usually older) will specifically state the printing on the copyright page (e.g., "7th printing"), but most instead have a series of numbers such as "11 10 9 8 7 6 5". The last number in the sequence indicates the printing, thus the above sequence would indicate a 5th printing.

[rivka, Librarian Moderator]

Good point. That would make sense.

I wonder how that pertains to the original question . . .

[John]

First, thanks to all (Rivka, Seizure Romero, and Michael) for your input. From the Wikipedia article referenced by Seizure Romero I found (among other interesting information about ISBNs) this website: http://isbndb.com/ . It appears to be a good source for identifying books via their ISBN, although Seizure Romero's warning that without the book in hand, information about it must be treated with caution. Even if a scan of the copyright page is available, one cannot be absolutely certain that the scan was from the correct book.

Pshaw, my better half is calling me to go for a run. I'll post a bit more after we're back.

[John]

OK, I'm back. Here are the conclusions I draw from the input on this thread.

(1) ISBNs are not back-assigned to an edition of a book that was originally published without an ISBN.

(2) Except that a book edition where the first printing was done before ISBNs were available and subsequent printings of that same edition extend into the time when ISBNs are available, may have an ISBN for the later printings (even though the later printings are not a new edition).

(3) We (GR librarians) should treat with caution ISBN/edition information obtained from any source other than the book itself.

Here is the lesson I learn from this: If a book printed before ISBNs were available is said to have an ISBN, this is an error. It is unclear whether the error is the ISBN or the date of the printing. Therefore a correction shouldn't be made unless reliable information on the nature of the error (preferably the book itself) is at hand.

I have to admit that this is great fun. I've learned a lot since engaging in GR librarian work, and I hope to learn even more in the future.