Hal Cain wrote:
And there are others which haven't survived at all, at least not in recognizable form ... In the meantime people write about them, or produce editions of works created to express opposition, and we have to formulate headings (citations, whatever) to deal with them in providing access.
Also, imaginary works have to be accounted for when it comes to subject headings, just like for fictitious characters (for example, LC has established subject (not personal name!) headings for Lynley and Havers in Elizabeth George's mysteries). One can imagine expressions of imaginary works, as well as fictitious manifestations and items.
... Part or whole? RDA seems to follow the implied reliance of AACR2 on usage: how is the work, part or separate, primarily known?
That was the important question in the heyday of the main entry! And it made economic sense of course, limiting the number of cards. But even then, catalogers couldn't possibly establish the facts in all instances. Today, of course, "primarily known" doesn't mean other titles should fall by the wayside! Cartainly not titles of parts when these can be cited as individual titles in their own right. RDA, like AACR, still leaves all options open, and so no improvement over the annoyingly antiquated "contents note" will be achieved, disregarding the noble intentions of FRBR. B.Eversberg

