Citing FamilySearch Databases vs Catalogs/Collections

Oh, what a tangled web we weave

When we set out to read

Manuscript materials digitized, processed, and published online!

In search of reliable answers to your questions, I'm going to walk back through what you've done and think out loud as I'm doing it.

First, let’s address the issue of what to use as a “title” to a set of records—cataloging data or …?  In EE’s discussion of FHL’s imaged materials, you’ll find this:

EE 2.27

To create our citation, we should copy precisely the label from each filmed book or file. If a register or file is not labeled, we should look for the target that the GSU imagers placed at the start of the material. If the target is missing or seems to be inaccurate, consult the catalog entry and create a generic label (record type and time period) using the descriptive note for that film. Be aware, however, that the FHL catalog description frequently uses a generic label to describe the contents of an entire roll. The actual title of a specific register or file may not appear in the cataloging entry.

Re the first boldfaced passage:

We copy the label from each filmed book or each filmed file--not the label that the FHL catalog puts on the whole roll of film, because a roll of film often contains several different items. The set of words that FHL’s catalog uses for the whole roll is not considered a "title," but a “descriptor.”

Re the second boldfaced passage dealing with cataloging data vs. specific content used:

In your case, when we click back to the start of the roll (image 3) to see what identifiers are given, we do find a target. It tells us this:

TARGET DATA:

Title of Record: Documents of Estates, 1-2671

Locality of Record: Natal, South Africa

Location of record set when filmed: State Archives, Pietermaritzburg

The film target does not identify the agency in Natal which created that record. FHL’s cataloging data does provide that information, as well as some differing “labels”:

 CATALOGING DATA:

                Title of Record: Deceased estates, 1846-1940

                Author: Master of the Supreme Court, Pietermaritzburg

                References: South Africa, Pietermaritzburg estate file, 1846-1950

Still, nowhere, in either locale, do we see the database title that you give: “South Africa, Pietermaritzburg Estate Files 1846-1950.” We also note the absence of two other information statements that many FamilySearch databases carry:

• There is no suggested citation in a window at the bottom of the image.
• There is no collection title > path  shown at the top of the screen.

As best I can tell, that database title “South Africa, Pietermaritzburg Estate Files 1846-1950” is discoverable only through two other approaches:

1. At FS’s main search page, we query for Saker’s name. There the “hit” for the death notice takes us to FS’s record extract, where we find a suggested citation that does include the FS database title you are using.

"South Africa, Pietermaritzburg Estate Files 1846-1950," database, FamilySearch (https://familysearch.org/ark:/61903/1:1:QLR2-KXLY : 17 March 2018), Harold Lamonby Saker in entry for MM9.1.1/QLR2-KXLR:, 1917; citing Probate, Pietermaritzburg Archives (Formerly Natal State Archives), South Africa.

2. At the main search page, instead of querying for the name, we use the “Find a Collection” search box. That takes us directly to the collection, where we can then query for the name.

No wonder we’re all confused!

Here we might use two approaches:

1. Just copy what FamilySearch offers. However, that omits some critical detail, in EE’s opinion.
2. Create our own, as you did, with a careful effort to distinguish between

Layer 1: The details needed to identify the original.

Layer 2: The details needed to identify the database.

You’ve done well with creating your own. There’s just one bit of blurring. In Layer 1 (the data for the original) you use FHS' label as a title for the record and place it into quotation marks (“Deceased Estates, 1846-1950”), thereby making it an "official" or "formal" title.  If someone were to go to the archives at Natal and seek that record by searching for a record set with that exact title, would they find a collection there with that exact name? Apparently not. When I use the search form at NAAIRS (http://www.national.archsrch.gov.za/sm300cv/smws/sm300gi?20190611191254064D7968%26DB%3DNABE) I get the following results

• Query for that phrase as an exact title: no results.
• Query for “Estates”: no results.
• Query for Saker’s name: I’m given a database entry for him saying that he appears in Depot “NAB.”  Source“MSCE,” Type “LEER,” volume 00, System 1, reference 926/1948. There is no hotlink to take me to the record so I can analyze more deeply.
• When I look up all those acronyms on the NAAIRS site, none of them lead to a record collection called “Deceased Estates, 1846-1950”

In short: That phrase seems to be a generic construction by FamilySearch. Therefore, if we put it in quotation marks as an exact quote of a collection title, then it should appear in FS's Layer 2, not the original's Layer 1.

All of this leaves us with no exact collection title for the original that’s imaged.  Without an exact collection title for the original, we are left to devise our own description for the title field—in which case we don’t put quotation marks around that generic description. EE 2.22 and 2.62 guide us there.

Missing details

Above, in discussing FamilySearch’s suggested citation, I mentioned that it omits data EE would consider essential for this set of records.  Specifically, FS (and you)

• cite the one exact document
• cite it to just one image

Given that this one document is a “death notice,” which is normally a standalone record, the reader of your citation is likely to assume that the one document is the only relevant item for Harold Lamonby Saker. To the contrary, the file is 10 pages long (plus cover), with much other data. The citation needs to make this clear, so that users know to look for more.

All things considered, EE’s citation would tweak yours only lightly, to cite the larger file, after which the specific document is stated. The three layers below are visually separated by color:

Master of the Supreme Court (Pietermaritzburg, Natal), Deceased Estates, 1846-1950, file 926/1948: Harold Lamonby Saker, specifically “Death Notice”; consulted as "South Africa, Pietermaritzburg Estate Files 1846-1950," database with images, FamilySearch (https://www.familysearch.org/ : accessed 20 May 2019) > microfilm 1,768,608 > images 967 through 977 of 2407, specifically image 968; citing "Natal Archives Depot, Pietermaritzburg."

Yes, Sarah. That works, but read on ...

> Your citation highlights the fact that the "death notice" is the thing being specifically viewed, but I'm wondering whether we just leave that off and cite the estate file as a whole...

There are times when we would cite a whole file. But if we make a specific assertion, the citation we attach  in support of that citation, should cite the specific document that makes that assertion. If the document is multiple pages, we would also want to cite the specific page.  An analogy might help to explain the reasoning here: If you pull a piece of information from a book, would you just cite the whole book, without giving the specific page or chart or table, on the premise that a reader (or you at a later date when your recollection has gone cold) can just search the whole book to find the context from which you took that information?]

>the province of Natal no longer exists - it has been amalgamated into a new province called KwaZulu-Natal. Does this need to be expressed in the citation, perhaps using square editorial brackets? Or is it enough to use the "old" wording to reflect the creation era of the records?

Yes, EE would put the new locale in editorial brackets, saying something such as [now KwaZulu-Natal].