QUESTION: Is a new edition of a book (or other resource) a new work, an expression, a manifestation, or what?
SHORT ANSWER: A publisher's statement about the edition is considered to be manifestation-level information. If a text has been revised, updated, corrected, or expanded, or if the title has changed, it is likely to be considered a new expression. If the text has been adapted or revised to the point where the nature or content of the work has substantially changed, it is considered a new work.
LONG ANSWER: Generally speaking, a publisher’s statement about the edition is considered to be manifestation-level information. In Tracey's workshop example "Do androids dream of electric sheep?" as you can see on her cheat sheet released in Q-Tip #6, the edition statement "1st Ballantine Books trade pbk. ed." is assigned the entity level of Manifestation.
There are metaphysical ways to look at the WEMIness of this question, but here are two crude, practical ways to figure it out. First, as you'll hear next week in Jean's workshop about the structure of RDA, RDA is arranged so that information about Manifestations and Items is presented in Section 1, chapters 1-4, and information about Works and Expressions is given in Section 2, chapters 5-7. All the familiar AACR2 rules about recording edition statements appear in RDA in chapter 2, at 2.5, which is a chapter on Manifestations and Items. Therefore, edition statements must be manifestation-level. And second, remember that you never see Works and Expressions walking around live in the bibliographic jungle. Works and Expressions are abstract concepts. If you are holding an actual, physical edition in your hand, it must be a Manifestation.
As catalogers immediately noted in Tracey's workshop, however, WEMI can be hard to nail down. Tracey mentioned the idea that WEMI exists along a continuum; Jean has a splendid full-color diagram of this continuum for you in her workshop next week. When would a new edition be a new Expression? Could a new edition be a new Work? RDA has some very specific guidelines about how to handle these slippery situations.
It is possible for a new edition (using the word "edition" as a loose idea, not necessarily as a classic 250 statement) to be a new expression. Tracey's presentation used as an example "Recreational problems in geometric dissections" by Harry Lindgren, which described itself as a "revision of Geometric dissections." Tracey's pop quiz and the original FRBR document written by IFLA assigned the entity level of this content in WEMI to Expression. The section of RDA that tells us how to handle this situation is in chapter 6, a part of RDA on Works and Expressions. 6.27.1.5 says:
If the work is presented simply as an edition of the previously existing work, treat it as an expression of that work (i.e., use the authorized access point representing the previously existing work). If it is considered important to identify the particular expression, construct an authorized access point representing the expression as instructed under 6.27.3.
This instruction is the rough RDA equivalent of AACR2 25.2B ("Do not use a uniform title for a manifestation of a work in the same language that is a revision or updating of the original work. Relate editions not connected by uniform titles by giving the title of the earlier edition in a note in the entry for the later edition and by making an added entry as appropriate") with the "authorized access point" in RDA-speak being the AACR2 "added entry."
Here's another example of this situation, to add to Tracey's example. It's at RDA 26.1.1.3. The resource described is:
Roget's Thesaurus of English words and phrases. — New edition / completely revised and modernized by Robert A. Dutch.
And the Authorized Access Point Representing the Related Expression:
Revision of: Roget, Peter Mark, 1779--1869. Thesaurus of English words and phrases.
While the publisher’s statement “New edition” is at the manifestation level, "Revision of" is given as an example of "Derivative Expression Relationships" in RDA J.3.2:
revision of An expression of a work used as the basis for an updated, corrected, or expanded version [metaserv:emphasis added].
As Glen correctly pointed out in Tracey's presentation, however, it is possible to consider something like "Recreational problems in geometric dissections" as a new Work---if, after careful examination of the content, it is considered sufficiently different from the previously existing work. How to determine whether is it sufficiently different? RDA 6.27.1.5 gives this guidance:
Adaptations and Revisions
If one person, family, or corporate body is responsible for an adaptation or revision of a previously existing work that substantially changes the nature and content of that work [metaserv:emphasis added], and the adaptation or revision is presented as the work of that person, family, or body, construct the authorized access point representing the new work by combining (in this order):
a) the authorized access point representing the person, family, or body responsible for the adaptation or revision … and
b) the preferred title for the adaptation or revision
There was some discussion in the workshop of Tracey's example of a new work:
Pride and prejudice and zombies : the graphic novel / Jane Austen and Seth Grahame-Smith.
Jane Austen is named first in the statement of responsibility and according to Peter Martinez contributed 80% of the words of the text. Why doesn't Jane Austen get main entry, to use AACR2 vocabulary? Because, darn it, when you add zombies to the text, you have substantially changed the nature, if not the content, of that work!
All clear now? Don't worry if it isn't. We have lifetime of practice in WEMI ahead of us.