[I am deleting a lot of the earlier discussion in the thread, to keep article size down]
Again, you are confusing the Preferred Title with the Authorized Access Point. The Preferred Title does not have to be distinguished from other Preferred Titles. It is the rules for Authorized Access Points (6.27) which detail how to add qualifiers to Preferred Titles to create AAPs. Preferred Titles themselves do not include qualifiers, ever. Preferred Titles can conflict with other Preferred Titles.
I will assume you are actually asking about Authorized Access Points, rather than Preferred Title. I do not have this particular title in hand. But based on what I know so far, my first instinct would be to not even bother making a Work NAR for the collection—the same decision as you find in OCLC #809833452. If I have no reason to believe that any of the short stories in the collection have previously been published separately, there is really no need to create an NAR for the Work, just as there is no need to create NARs for most monographs. That may change in the future as we move toward Bibframe, but right now, I don’t need to create an NAR. The Preferred Title would be the 245$a field, and the AAP would be the 100 plus 245$a.
If for some reason I did choose to create a NAR for this title, I would be inclined to choose Negras as the Preferred title. The Work AAP would then be Arroyo Pizarro, Yolanda. Negras. Again, there is no reason to believe these stories have been published separately, so their entire existence is part of this collective Work. The component stories have not been compiled from other sources; there is no reason not to treat it as a coherent whole.
If one or more of the stories had been published previously in a different collection or singly, I would check whether this is the first time this set of stories have been published as a compilation. If the compilation has had other names, then I would almost certainly use a conventional collective title. Next I would ask myself, “If a reference source talks about this collection, would they use a particular name for it? If they do, would they be talking about this particular Expression, or about a conceptual collection of these stories, i.e. the compilation as a Work?” If I decide that the collection does have a name that it is known by _as a Work_, I would use that as my Preferred Title and base my AAP on that title. As I noted above, if this is a collection of original stories not previously published, I would probably decide that this collection would continue to be known by that name.
If I choose to use a conventional collective title, and the author has written other things besides short stories, then the Preferred Title would be Short Stories. Selections. The AAP would then be Arroyo Pizarro, Yolanda. Short Stories. Selections, probably with a parenthetical qualifier. (The LC-PCC PS says to not predict conflict, but I probably would add a qualifier even if there were no other collections by this author with that conventional collective title). I could use Negras as the parenthetical qualifier, but I would be more likely to use the date. There are numerous other possible qualifiers.
If I chose Negras as the qualifier, there would still be no conflict with a short story by that same name. The Work AAP for the collection would be “Arroyo Pizarro, Yolanda. Short Stories. Selections (Negras)”, while the AAP for the short story would be “Arroyo Pizarro, Yolanda. Negras.” They are clearly distinct AAPs.
As for your question about Orestea: if the 240 field is “Orestea”, it represents part of the AAP for the compilation; the full AAP is the 100 plus 240. Since the 240 does not include qualifiers, it also happens to represent the Preferred Title of the compilation. If the 240 were “Godzilla (1989)”, then the Preferred Title is “Godzilla”, without the qualifier. MARC is not designed for RDA. RDA requires both the AAP and the Preferred Title to be recorded, but MARC does not have separate fields for Preferred Title and AAP. Instead, the 1xx plus 240 or 245 represent the AAP, and the Preferred Title is merely part of the 240 or 245.
Yes, we were discussing “how to construct a Preferred Title.” I used the Name/Title AAPs for the purpose of illustration. When “Orestea” is in 240 field in the bib, does it represent preferred title of the compilation?
As for qualifiers, if “Les negras” also happened to be the title of a short story in the compilation, what would you choose for the “preferred title” for the compilation? How would you distinguish one from the other?
You are confusing the Preferred Title with the Authorized Access Point. We were discussing 126.96.36.199, which describes how to construct a Preferred Title. Preferred Titles do not include distinguishing qualifiers—those are added when constructing Authorized Access Points. See 6.27 for the rules for constructing Authorized Access Points from Preferred Titles.
Yes, to be frank, I am reading 188.8.131.52.3 basically the same way. It is difficult for me to regard a compilation like “Las Negras” as a having a “commonly known title” in the same way as I think of the trilogy of Orestea by Aeschylus. The title of the latter can be found in all sorts of reference books and other resources. It is therefore essential to establish the Name/Title as “Aeschylus. ‡t Oresteia” (ARN4417723=LCCN no 97050641). Various compilations of poems by Horace, such as Odes, Epodes, Satires (Sermones), Epistles, have “less commonly known” as well as “commonly known” titles, for which, obviously, preferred titles must be established. So, I do use the first instruction of 184.108.40.206 when cataloging such compilations, taking into account the long history of textual transmission.
LC/PCC instruction for 220.127.116.11.3 provides us a guideline for treating a compilation of “more than two works” which I find both useful and logical, but somewhat limiting in that it does not seem to encourage the use of “parenthetical qualifiers.” In an earlier message, I expressed my preference to using such a qualifier: 100 1# Arroyo Pizarro, Yolanda. $t Short stories. $k Selections (Golpes de gracia) (ARN10333589 =LCCN no2015159874). The original title Golpes de gracia is still there and is easily searchable in the bib (245) and in the authority file (400). The advantage of doing this, instead of a generic “Short stories. Selections,” is multifaceted and far-reaching, as I mentioned previously. So, I don’t see why we cannot construct a preferred title like 100 1# Arroyo Pizarro, Yolanda. $t Short stories. $k Selections (Las negras), instead of a nondescript Arroyo Pizarro, Yolanda. $t Negras, the former being more descriptive and precise than the latter.
English cataloging institutions would benefit from having this type of conventional collective titles when dealing with non-Latin materials (Arabic, Persian, Hebrew, CJK, etc.). From a user’s perspective, when doing a Name/Title search under a specific author, one could clearly identify at first sight what these compilations are. For catalogers, it means that each new compilation would have to have a new N/T authority record. More NACO work, to be sure, but more benefit for all. That was the intent of my original message.