Separate item type for published letter

Although Zotero's difficulty in dealing with this type of citation has been mentioned in previous posts, some dating as far back as 2007 ( , seventh contribution) I'm creating a separate thread because the full scope of the difficulty has not yet been described. For this reason the developers remain sufficiently unaware of its contours that they can dismiss it as being largely addressed by the "book section" item type (see

The "book section" type is in fact terrible at generating a bibliographic note for a published letter, at least in Chicago style. For example, I'm citing a 1621 letter (month & day info didn't survive) published in a book entitled Seventeenth-century Economic Documents. The only way to enter this type of item (as a Book Section) in such manner that all the info will be generated in the note is to put the sender, recipient, and date in the Title field. But then a number of manual corrections are required. Here's what Zotero generates for a full note...

Joan Thirsk and J. P. Cooper, eds., “G. Manners et. Al. to Privy Council, 1621,” in Seventeenth-Century Economic Documents (Oxford: Clarendon Press, 1972), 608–9.

...yet CMoS requires that the citation be formatted as follows:

G. Manners et. al. to Privy Council, 1621, Seventeenth-Century Economic Documents, eds. Joan Thirsk and J. P. Cooper (Oxford: Clarendon Press, 1972), 608-9.

Here are the manual corrections the user must make:
* delete the comma from "eds.,"
* in sender-recipient, replace the capital A in "Al." (which Zotero generates despite the fact that it's lowercased in the actual database Title field) with the original lowercase version of the letter
* delete quotation marks around sender-recipient
* delete the "in" that comes after sender-recipient
* cut "eds." out of its current place
* paste "eds." at the front of the editor names
* delete comma after editor names
* type comma after book title
* cut "eds." & editor names from their Zoteroed position
* finally, paste "eds." & editor names to their proper place after sender-recipient

That's ten, count 'em, ten manual adjustments that need to be made, an unacceptable level of work for a fairly common type of citation; if you're distracted by an important or lengthy phone call at any step in the process--or if it's the end of the day, you're tired, and you just want to finish your current page and get the #$*@ home--it's a good bet you'll inadvertently skip one or two of these niggling operations. Published letters clearly need their own Item Type. Ultimately two will be necessary, the second one for letters published in journals, but those are far less common than letters published in books and can be ignored for now.
  • Hmm - I'd really want to avoid an item type for "published letter" - even if we get this in Zotero, I can pretty much guarantee you that it's too specific as a category to make it into the the Citation Style Language (and then it wouldn't be much help). Remember that Zotero is used by people from all types of disciplines and the only way to keep item types somehow manageable is to keep them as abstract as possible. (As a political scientist, e.g. I think that reports, working papers, UN Resolutions, International Treaties, EU directives - to name just a few - are all very distinct item types. You see where this is going as you add a musicologist, an archaeologist, and a performance studies person in the mix.

    I can think of two solutions that don't require separate item types:
    1. We could add publisher/editor information to the letter item type. Since the general structure for citing published and unpublished letters is quite similar, that would seem to work quite well. Still requires Zotero changes, but those would be pretty easy to implement with the planned field-update in 4.2.

    2. We _could_ actually do this in CMoS right now by simply reconfiguring how the style behaves when there is no author for a book section. Right now it uses the editor instead, but we could tell the style to use the title without quotation marks - leaving the editors in their normal place - and that looks like it'd come out just right.

    Any thoughts?
  • We could add publisher/editor information to the letter item type. Since the general structure for citing published and unpublished letters is quite similar, that would seem to work quite well. Still requires Zotero changes, but those would be pretty easy to implement with the planned field-update in 4.2.
    IMO, this is the way to go. Zotero users should not need to figure out that a letter should be cited as a book section.
  • for 1.) which fields would we need?
    Add editor & translator as additional creator types,
    add book title (label?), place published, and date published - anything else
  • Instead of book title, I think it would make sense to go with a more generic "publication title" so that it covers journal publications also. I think this will also call for volume/issue. Idk if ISBN/ISSN would ever be required. DOI will probably get added to most types anyway.

    Having more fields (as long as we can map them to CSL) cannot hurt. if they're not applicable, they will just be ignored.
  • OP seconds aurimas's suggestions, including generic "pub title" which would also take care of journals. (and do call it "title" rather than "label".) adamsmith didn't mention the press's name along w/pub date, so make sure to include that--all the standard book AND journal fields, e.g. the volume fields, will have to be added, w/some kind of IF statement (or other programming structure) making book-only fields mutually exclusive from journal-only fields. Also you'll have to make sure to differentiate shared fields such as Date from fields that appear to be shared but are not, such as Volume. By way of example, the latter might necessitate having a "book volume" field separate from a "journal volume" field, since they must be treated so differently in the output, although it may also be possible to take care of that with an IF during the output as well, but there'll have to be some kind of switch. An entry in the "publication title" will serve as a switch letting Zotero know that it's not a manuscript, but that still leaves it unclear at output time whether it's a book or a journal.

    This is the kind of difficulty that led me to propose separate item types--putting letters all into a single type is definitely better for users but will make it trickier for developers. (disclaimer: I can't write serious code--know just barely enough self-taught javascript to sling pop-up windows around a web page.)

    Can't think of anything else (except I'd love it if DOI could simply be abolished--I don't know what happens with other styles, but for Chicago-style output, Zotero actually generates the DOI as part of a full note, so you always have to either delete it manually from your text, or you have to delete it from the Zotero item itself--for obvious reasons I do the latter, which makes DOI completely useless.)
  • Chicago Manual strongly encourages inclusion of DOIs where available. That's why they get included. It's trivially easy to remove them from the style.
  • Interesting--must be a fairly new recommendation. (This dates me more than I'd like, but I own the 14th ed.) Nobody in my field is including DOIs in citations yet even though Chicago is the by far the most prevalent style.
  • It was added for the 16th edition, yes. I actually think it's a great requirement and it's a shame that academics are so reluctant to take up DOIs.
  • edited March 30, 2018
    Has this ever been implemented?

    I'm using Zotero 5.0.41 and have to cite an individual letter which was published in a collected works volume. The CMS 17th ed. (15.43) says for author-date use the date of the collection and "weave" the individual correspondence into the text. But even if Zotero could do this, the job of keeping track of relevant individual correspondence requires a different approach.

    Therefore an expanded version of a book section, as was discussed on Oct. 4, 2013, might be the way to go. I see no sign of this. Any progress here?
Sign In or Register to comment.