Sources from the Danish National Archives

Lene:

First one point should be made clear: I am not a Danish researcher. That said, we can discuss the principles of citation and then you, as the experienced Danish researcher, can make your own decisions as to what works best with the documents you are using. Ideally, other experienced Danish researchers will also weigh in on this discussion.

1. A citation typically begins with the identity of the creator of that source. This holds true whether we are dealing with an authored work, a published compilation, a governmental record, etc.

2. If the creator of a census is a "state" level government rather than a national government, then the state government is cited as the creator.

3. In addition to citing a government or agency in the "author/creator" field, our citation will also identify the archive or government agency that holds the document.

Given that "arkivskaber" means archive creator, the "categorization" given by the Danish National Archives is ambiguous to anyone who is not familiar with the history of Danish censuses. Does the DNA mean to say that the creator of the census is the state-level archive? Or does it mean to say that the creator of the catalog entry that describes the series is the state-level archive?

Or to put it another way (given the structure of the search screen at the deep link you posted), does the identity of the arkivskaber enter into consideration as a citation element only as one bread-crumb or waypoint that's needed to locate the census in this particular database?

Lene,

For the benefit of forum readers who do not have the 2nd edition that you cite, I'll put the example here. (That particular QCM, p. 240, is not in the current edition.)

The source list entry does, as you say, start with the jurisdiction. However, the first reference note (which is the citation format we discussed in our first round above), does not start with the jurisdiction. Two differences exist here:

• Source List Entry: Here, the overriding need is to organize our source list efficiently, so that we can easily find a source when we need it. (See EE 2.50 Source List Arrangements: By Geographic Locale.) The U.S. government—more specifically, the United States, Bureau of the Census—is the creator of this record. Very few researchers want to organize censuses under that label. For U.S. censuses, they prefer to arrange their censuses by specific locale. Thus, the source list entry in this example begins with the locale. EE 2.51–2.53 also demonstrate other arrangements for source lists that some researchers prefer.
• First Reference Note: Here, the identity of the creator is typically shown in the first field. In this QuickCheck Model above, that first field identifies both the creator (the United States government) and the year. We could, of course, create a census citation that begins explicitly with the name of the creator and then identify the year. For example:

United States, 1850 census, Marion County, Iowa, population schedule, etc.

However, that creates an extra field for software developers to deal with and it places the country identification before the census year, while most users—who will be citing a country's census dozens or hundreds of times within their research project—prefer to have the year first. To resolve both issues, it has become standard practice to combine the year and the creator's identity into one field: e.g., "1850 U.S. census," or "1836 Denmark census."

Now for those state-level censuses: The p. 240 QC Model that you reference is not a state-level census. The state of Iowa is not the creator. The QC Model for state-level censuses, in the 2d edition, is on p. 252 "Microfilm: State-Sponsored Censuses." Other undiagrammed examples are also at EE 6.46. You will notice that the format for the state level follows the same format used for the national census, but the wording of the "Census ID" field indicates that it's a state-level census rather than a national-level census.

Your last paragraph makes an important point: Different repositories handle materials in different ways. Our citations have to include the necessary data to find a record in the repository where it's housed. That's one reason why EE's chapter on Fundamentals of Citation starts off (2.1) with "Citation is an art, not a science."

Thanks for educating us all about the ongoing changes at the Danish National Archives.

Lene, you can add anything to your citation that you want to add, if you think it will help you or others to analyze the reliability of the information or understand the record. Typically, comments are added in a separate sentence (or more) after the citation. When we add information from personal experience or information we have gotten from an archivist or someone else, it's best to also add how we know this information—who or where we got it from.

Use of waypoints, breadcrumbs, or paths in a citation to a complicated online database ...

Sue, you are likely correct that some users of your research will not understand the  >  symbol that is used to define the navigational path at websites such as this one. Realistically, this is a problem we face with every kind of citation. Newer researchers don't understand conventions until they learn them or until we explain them.  Ergo, we have two choices:

• We could, as you suggest, use a different symbol or punctuation—but that would also confuse all those people who do understand the use of the standard symbol. Odds are, too, whatever other symbol or punctuation we would choose, the inexperienced researcher would still be confused by all the information in the citation—especially the non-English words.
• We could use the standard symbol and see this as an opportunity to teach the inexperienced.

EE is a strong believer in following standard practices—"rules of the road," we might call them. When we are driving and want to turn left, we use the appropriate blinker or the appropriate turn signal. We don't use a different signal that makes good sense to us but—because it's not standard—would not be understood by most people.

You have suggested the use of commas to separate those different waypoints in the FS citation.  Specifically:

Soro amt, Slagelse herred, Sonderup parish, 1816-1852, Parish Register, image 68 of 533:

Here, we need to refer back to two points made at No. 2 in my message to you above. 

• By long-standing practice, we use semicolons to separate the major elements in a citation. Within each of those elements, we use commas to separate the individual parts of that larger element. This follows the same rules used in all writing: Items in a series, all on the same "level," are separated by commas, while items on the larger level are separated by semicolons. To apply this to the FS citation:
  • "Fødte, Konfirmerede, Viede, Døde, Til- og afgangslister, Register" are details all on the same level. They represent a register title that names similar items in a series: “Baptisms, Confirmations, Marriages, Deaths, Lists of Accession and Expuntions.”
  • “Soro amt, Slagelse herred, Sonderup parish, 1816-1852, Parish Register, image 68 of 533” are not all details on the same level. The first three are locations, at different jurisdictional levels; the next is a time frame; the next is a manuscript volume; the next is an image at a website.
• Also by long-standing practice, U.S.-style citations and U.S.-style references to locations both start with the smallest and work up to the largest. However, the series of items that you separate with a comma do the reverse. You obviously used that arrangement because it's the arrangement used by FS in organizing its files; but peas and apples are being mixed here, without giving readers an explanation of what you're trying to do.

The elimination of the path and rewording of the waypoints raises another issue. A citation to an online database has two or three basic parts:  

1. the identification of the database;
2. the identification of imaged material or item of interest;
3. the identification of the source, as given by the website. 

When we create a citation, we must be careful not to take details that belong to one part and mix it into a different part.  The path provided by FS  …

Sorø > Slagelse > Sønderup > 3 (1816-1852) Fødte, Konfirmerede, Viede, Døde, Til- og afgangslister, Register > image 68 of 533

… represents the specific labels that FS has assigned to files within its database. This is the specific data needed to locate that particular register at the FS website.  Your rephrasing eliminates the “signals” that identify this phrase as a path within that database and it alters the “official” identification of each file or subfile within that path. If we want to clarify what those words mean or substitute our own description of the location of the church that  created the register, that’s fine to do; but it should be done in a different layer that is clearly our own addition. Or better, we could cite the path appropriately and simply add a sentence that explains to the uninitiated what the path symbol means.

Returning to your phrase:

Soro amt, Slagelse herred, Sonderup parish, 1816-1852, Parish Register, image 68 of 533:

If someone goes to Soro, to the town of Slagelse and the parish of Sonderup (or to the National Archives that houses the register today) they will not find there a register in which there is an “image 68 of 533.” That image number applies only to FHL film and the FS database.  When we need to create a reference to the original parish and its register, then our citation would refer to the register name or number (in this case, register 3) and the page number in that register.

Sue, in bibliographic citations, a period is used to separate each major element. Reference notes are written sentence style, with commas, parentheses, and semicolons separating the major elements. It seems silly, but there's a logical explanation for the differences. I've detailed that elsewhere in the forum, in answer to someone else, so I won't go into the whole spiel again here.