Citation Shortcuts: A Researcher's Needs vs. Recordkeeping Systems



19 January 2014

A researcher from Sweden, raises this issue on EE’s Facebook page. Actually, he sidled into the issue sideways, from the angle of “short citations.”1 But, as with everything in life, when we design shortcuts, it pays to consider the larger framework of the research needs we need to fill.

Sweden’s archives offer millions of historical documents online. Sweden is also one of those with a superbly organized archival system. Each register has an exact alpha-numeric ID. It may even have more than one. We could contact the archives, cite that code or number and they could retrieve the record. (Assuming we or someone else haven’t made a typo in that number, of course.)  

So, borrowing one of the several citations Magnus offered—one formatted in a slightly different manner from the entry above—let us ask:

Does Gunnilbo A-1 (1688-1724) work for you as a citation?  Does it fill your needs, as an analytical researcher?  More specifically:

  • If you were to use someone else’s work and they cited their source as “Gunnilbo A-1 (1688-1724)” or "Gunnilbo-C-2-1684-1706-Bild-520-sid-48," would you know what kind of source had provided the information?
  • Would you be able to form an evaluation of the reliability of the information that came from "Gunnilbo A-1 (1684-1724)" or C-2 ....? (Would you know, without additional research, that the one word given to you, Gunnilbo, signified a parish—or where it was located?)
  • Would you be wishing that the earlier researcher had taken about 27 seconds more and added a few other words to identify the register by name, not just by the recordkeepers' alphanumeric code? (Are you also wishing the researcher had added a few more words to identify the archives that holds the record?) 

Let’s take this What if  one step further. Let’s say that the earlier researcher realizes most people aren’t experts in the recordkeeping system of the Swedish Archives. Sure, if we’re going to do research in Swedish records, we should be. But none of us start out as an expert in anything. So Magnus has helpfully extended the short-form citation:

Gunnilbo A-1 (1684-1724), image 208, page 202 (AID:v72130.b208.2202,NAD:SE/ULA/10346)

Are you now saying Ah, yes! Now it makes sense!?

Probably not. What you’re actually thinking is more likely to be Darn! Now I have something else to learn before I can figure out what's been used here! In this example, the researcher has indeed pointed us to a learning tool, a guide we can seek in the National Archives Database (NAD)—once we find the URL, of course—a database in which “SE/ULA/10346 is a reference number to a page that describes where the books are stored in the Swedish Archive.”3  

Wonderful. If we’re serious researchers in any country, we should be reading archival catalogs to learn how and where records are stored. But, I’m also willing to bet that every reader at this point is thinking, Please, sir! Could you just give me a citation that tells me what we’re dealing with? Plain words? Something like

Gunnilbo (Västmanland, Sweden), “Household Records, 1688-1724,” A-1:202; Regional Archives, Uppsala, Sweden.

Of course, the extended alpha-numeric citation presents us with another problem. Why is it citing both image number and page number?  If the researcher used the original volume, the standard would be to cite volume, page  or vol:page.  The insertion of the word image suggests that we’re dealing with a digital copy, and that opens up another riff of issues, starting with Whose image? After all:

  • An image copy of a record, offered by one provider, can differ in content and quality from an image of the same record offered by a different provider.
  • Research standards, almost everywhere, call for identifying the creators of the materials we use. When images are accessed through a database at a website, we need to know at least (a) the name of the database; (b) the name of the website; and (c) the standard publication data (Place/URL : date).
  • The image number, which usually is usable only for that particular website's version of the register, should then be added after the publication data for the database, in the field one would use for a page number in a published book or a frame from a roll of microfilm.

All things considered: EE would be inclined to use this citation:

  First Reference Note:

Gunnilbo (Västmanlands län, Västmanland, Sweden), “Household Records, 1688-1724,” A-1:202; Regional Archives, Uppsala; digital images, “Swedish Church Records,” ArkivDigital ( :  19 January 2014), for Gunnilbo A-1 (also numbered 72130), image 208.

Subsequent Note (the shortened citation):

Gunnilbo, “Household Records, 1688-1724,” A-1:202.

Agree? Yes? No?



        1.  Magnus Sälgö, comment, "Evidence Explained," Facebook ( : posted 11 January 2014).

       2. Magnus's posting also pointed to Salgo60, "Importerat Släkträd 2010 Aug 15," Ancestry (|aid%2Cpid|cid : accessed 18 January 2014), attachment to individual page for "Johan Adamsson Stockhaus," which provides the nice graphic for today's blog posting.

       3. Sälgö, comment, "Evidence Explained," Facebook, 11 January 2014.

salgo60's picture

I have started to play around in nonSwedish land Wikitree and start realizing how much more difficult the world of citation is when someone outside the Swedish genealogy interest group should understand how we quote the Swedish church books

Thanks for your answer and Regards from Stockholm, Sweden

geogenie's picture

I have used ArkivDigital for a few years and am now trying to create citations for records. Starting with what you have here and what is also in EE 7.45 I have developed an alternative and am interested in any feedback on it.

The digital images are actually obtained through the ArkivDigital online software, not on the website (which provides a link but not the image), which is why I added "online." I removed "Swedish church records" as the title. Not all the records available are church records (some are court or police records, etc.).

I also removed the dash in the volume name and replaced with the colon, which is the way it is usually shown. Some volumes in other locations actually have two letters like AI:2.

I also added the specific item on interest on the page (the baptism) so we know what entry the citation is about.

The AID number is the shortcut used in the software to go directly to the image so I felt that it was important to include to make it easy for anyone to find. Only people who have the subscription can see the images so they should be familiar with the AID.

One question I have though is should we use the Swedish names for the records and the locations or is the English description acceptable? I did not put the name of the records in quotations since I used an English translation.

Please review:


Gunnilbo Parish (Västmanland County, Sweden). Birth and baptism records, 1684-1706. Regional Archives, Uppsala. Digital images. ArkivDigital Online. : 2016.

First Note:

Gunnilbo Parish (Västmanland County, Sweden), Birth and baptism records, 1684-1706, vol. C:2, p. 48, baptism of Johan Adamsson Stockhaus; Regional Archives, Uppsala; digital images, ArkivDigital Online ( 19 June 2016), for Gunnilbo vol. C:2, image 520 (AID:v72162a.b520.s48).

Subsequent Note:

Gunnilbo Parish, Birth and baptism records, 1684-1706, vol. C:2, p. 48, baptism of Johan Adamsson Stockhaus.

Any suggestions would be helpful! Thanks for reading.



EE's picture


As you probably know, I posted a notice about your query on EE's Facebook page to alert other experienced users of ArkivDigital's software, in hopes that one of them would weigh in. We had a couple of responses there, but they did not address your issues. Below, I'll do so in the order in which you raised them.

Elimination of "Swedish Church Records" from the citation:

Two issues here:

  • "Swedish Church Records," which our citation puts in quotation marks, represents the title of the database that was used. Naturally, we don't use those words if we use a database or a collection that has a different title.  By eliminating this field, you are implying that within the "software" that you have to download, there are no separate databases or collections—that all records are mixed together into one single entity. Is this what you intended?
  • If you feel it is essential to note that the record is delivered through a special software, would you not want to actually say that? The insertion of the word "online" will not be interpreted that way by readers unfamiliar with the system you are using.

Dash vs. colon

You point out that you have "removed the dash in the volume name and replaced with the colon, which is the way it is usually shown. Some volumes in other locations actually have two letters like A1:2."

Dashes, colons, and diddly dots are important only to the extent that they enhance or diminish clarity. Five issues seem at play here, with EE 2.63 being the key passage in the manual.

  • The longstanding convention, when citing a numbered volume, is to place a colon between the volume number and the page number;  2:48, for example, means volume 2 and page 12. 
  • In a series of manuscript registers, we will often see the register identified by a letter-number combo: sometimes C2 and sometimes C-2. . We may write those as C2 or C-2. When we add a page number, the result would  be C2:48 or C-2:48.
  • Whether a dash is used between the number and letter is usually insignificant. The significant point is that in standard citation language, what comes before the colon represents the volume and what comes after the colon represents the page.

Within this framework, there is also a variant.

  • In some very large series, there may be sub-series. For example, within Series C, there may be numbered volumes 1-99.  In that case, the pattern is Series:Subseries:Page
  • Within this standard framework, identifying a volume as C:2 (plus page number) carries a quite different meaning from C2 or C-2. The standard interpretation by most readers would be that you are referencing Series C, within which they should look for a Volume 2.

Our conundrum, as researchers and writers, is this:

  • Different archives, courthouses, and rectory offices across time and place have used all kinds of systems.
  • Within our own work product, we are expected to use a consistent format or pattern—regardless of the archive—so we will not confuse our own readers.
  • Consequently, if consistency requires us to make some variance in the way a record is identified in This Archives or That Archive, the expectation is that we will explain this in our first citation to the source.

Adding specific item of interest

The “specific item of interest” most often cited for a book or register is the page number. If there are multiple entries on a page and there is a need to specify just one of them to avoid confusion, then we may add the entry number or the specific names involved. If our text or narrative identifies the parties involved, then our citation need not repeat that information unless, say (a) it is not possible to locate the specific entry without it; or (b) the entry misspells the name in a way that might not be recognized. In the latter case, our citation would use the exact spelling in quotation marks.  That said, adding to a citation the name of a child being baptized, along with the page number, does no harm. Some people prefer the explicitness. Some people value conciseness. EE demonstrates both.

Using original Swedish title or an English translation

EE 2.23 covers this: “Citing titles in Foreign languages.”

The Editor

geogenie's picture
Thank you for the detailed

Thank you for the detailed reply. I think I am mostly stuck on how to address the use of the ArkivDigital (AD) software. I suppose there is a database behind the scenes which they refer to as the "Archives." There is a second database called "Index" which is for a population search, which I am not citing in this example.

I'm linking to some screenshots so you and anyone else can see what it looks like.

The first image shows the home screen for the archives search in the software. You can enter a parish name or authority that created records or filter by several options. There is no search that leads directly to the images.

The second image shows the results after clicking on Gunnilbo. This is the list of all the volumes available for that parish. Once you click on one of the results, then the entire book in digital format is available to browse.

If you know the AID, you can enter it in the box at the top left and go directly to the image.

For the dash vs. colon, you can see in the images what the convention is for AD (which was developed by Sweden for their archives). I was trying to maintain this convention and was concerned that since they already use a colon in the volume, then adding a colon between the volume and page number would be confusing (C:2:48). I didn't know earlier why it would be confusing but you explained that above. I understand the point now that there should be consistency in citations between the different systems used by different repositories.

For the specific item, that does make sense now that I read that. The footnote will be referencing someone in the narrative so it may not be needed.

Given all this, here is a new attempt. Does the software reference work?

Gunnilbo Parish (Västmanland County, Sweden), Birth and baptism records, 1684-1706, C-2:48; Regional Archives, Uppsala; digital images, "Archives," ArkivDigital ( 19 June 2016), for Gunnilbo C-2, image 520 (AID:v72162a.b520.s48), image accessed through ArkivDigital Online software.