Chinese references in text

Hello,

I have the following problem when working with Chinese sources: I fill out the Zotero mask with last name in Latin (let's say "Wang") and first name the Latin version + the Chinese characters (i.e. "Wei 王伟). In this way my bibliography looks good as it shows: "Wang, Wei 王伟 (2010), etc...". But when I insert the reference via Zotero into my text, the system shows the characters as well i.e. "W. 王伟 Wang 2010". I understand that Zotero does that to differentiate this Wang 2010 fom all the other Wang 2010 in the text. But I can't have the characters. When I delete the characters by hand I am informed that this reference will not be updated when I change the citation style etc. So, is there a better solution than that? Is there a way zotero will only refer to the Latin part of the name? Maybe a third field in the mask of zotero standalone? I downloaded Juris-M but couldn't find a solution here either. As I am facing more than 400 pages of doctoral dissertation I would be very thankful for a feasable solution. If there is another thread having found one, please lead me to it. I couldn't find it when searching through the forum. Thank you!

Kind regards

Ryanne
  • I understand your question, but I'm not actually sure of the best answer. I just wanted to verify that it makes sense, and to share a couple thoughts about possible workarounds.

    I do something similar with Chinese names, but I put them in brackets. So the author entry in Zotero would be like Wang, Wei [王伟]. If you add those brackets, does it help at all? (I'm actually not sure what sort of processing Zotero does behind-the-scenes with that format for input, but you could test it in case it is helpful.)

    But for me, I'm using a different kind of style, where instead of adding first names, letters are added to the year. So instead of "W. Wang 2010" and "X. Wang 2010" to disambiguate, the styles I typically use would produce "Wang 2010a" and "Wang 2010b" (yes, even if they're different people). So that's why I'm not sure about the solution to your question, with the styles you're using: if I used my library with one of those styles it might end up with the same problems.

    Another option for you, if you're allowed some flexibility, would simply be to use a different style that disambiguates differently. Or you could create a custom variant of your current style and just change the disambiguation method. That's very easy to do (changing one parameter in one line of code!). It would NOT be exactly the same as the current style, so it might be "wrong" if there are strict requirements for the formatting of your thesis. But if you have some flexibility, that would be the easiest answer. You could either only add initials for disambiguation (I think this would just do first initials, and not the characters, unless the first initials were still ambiguous!), or you could set it to only add letters to the year instead. There's also a setting for whether you want disambiguation to apply in cases where author names are different in general, or only in cases of actually having two papers cited from the same year that would actually be ambiguous. (So "Wang 2009" and "Want 2010" could be different people, but you wouldn't care unless they both wrote a paper in 2010.) That's all explained in the documentation for disambiguation. Somewhat technical, but you could play with the options to find what's best. Someone can help you if needed.

    If you need a manual workaround, then you should do everything you can to avoid manually editing active Zotero fields. Just don't do that. (It will cause various problems, as you've noticed.) One alternative would be to just let it format it like this. Then at the end, when you're done with your dissertation, save a copy, and unlink the Zotero fields so that you can edit the final version manually. Although it's 400 pages, it would only take you a few moments per page to visually scan for Chinese characters among the Latin alphabet. Just go through every citation and remove them. That would take maybe a few hours, but you'd only need to do it once. Or you could also more systematically do this with a search. Unless you're citing hundreds of authors, you could just search for each name (in Chinese characters) individually and remove them.

    Finally, and perhaps as the best advice, check out Juris-M, which is an expanded variant of Zotero (it exists in parallel, and you can use your existing library, no need to start over). It's specifically designed to handle multi-lingual situations like this, so it could directly fix your problem.

    (Sorry for the very long response here, but I thought I'd share my own perspective on having some some similar things. I hope that at least one of those ideas is helpful.)
  • Yeah, if that's viable, I think working with a style that doesn't disambiguate with initials is a smart workaround for this -- we can help you with that if you're otherwise using an existing style.

    For Juris-m, you'd have to enter the name twice, once as transliteration, once in the original and then you can specify in the preferences how you want those components to appear.
  • Dear djross3,

    thank you so much for your comperehnsive answer. I will first try to change the citation style :)

    Have a good evening

    Ryanne
  • (@adamsmith are Chinese characters treated as "initials"? Or as whole names? Would they always appear, if any (even initials-only) disambiguation is needed?)
  • I don't know, would have to test -- this isn't in the specs, and so citeproc-js is doing whatever Frank thinks is reasonable.
  • Dear both,

    I tried to change the citation style in the normal Zotero and later put the Chinese characters in brackets which both didn't work. Then I tried out Juris-M, but now I have have trouble with the bibliography. I have 2 problems so far: 1) I edited a Chinese source in Juris-M so I wouldn't have the Chinese characters in my in-text citation. The system thinks there are 2 titles from the same person and makes "Qu 2007a" and "Qu 2007b". I re-refered all Qu's to the Juris-M Library, but in the end I still have Qu 2007b although there is no Qu 2007a in the text anymore. In the bibliography there are still two sources. Why? 2) I add a Chinese source from Juris-M and it only shows the Latin version in the bibliography (no Chinese characters or translation). I looked into the handbook but don't understand. My language preference mask looks different then the handbook (or maybe I just cklick the wrong "prefences" button?). When I check the citation style preferences everything looks fine...
    Thanks again for your help.

    Ryanne
  • Oh, I forgot, there is another problem: I added a reference through the Juris-M library and it doesn't show at all in the bibliography. What do I wrong?
  • Before going into juris-m -- what was the trouble with changing the citation style? The above should definitely work in principal.
  • The idea was to tell the system to only use initials when quoting in-text, but it already did: the reference I got was "W. 王伟 Wang 2010", so I don't know what else to change.
  • No, the proposed idea was to _not_ use initials for disambiguation at all. As per djross3
    So instead of "W. Wang 2010" and "X. Wang 2010" to disambiguate, the styles I typically use would produce "Wang 2010a" and "Wang 2010b"
    (that's removing the disambiguate-add-givennames="true" attribute from citation in the CSL). If that's not an option, we can explore juris-m, but that's going to be a good bit more involved and there are fewer people with relevant expertise.)
  • Oh, sorry, no to have Wang 2010a and b isn't an option for me. My supervisor wouldn't agree and the publisher later neither, and I would have to change everything again. That's why I thought I try Juris-M.
  • (I haven't read the whole thread in detail, so if I'm missing something, let me know.)

    For information, citeproc-js will format text entered in (parentheses) or [brackets] following the rules for names overall—the first word before a space is formatted like a family/last name, subsequent words are treated like given/first names.

    In this case, I think that including the original Chinese characters in brackets or parentheses is probably the best option; conceptually, this is sort of like an alternate form of the name. It looks like you are using a style based on APA; in brackets would be APA's preferred format.
Sign In or Register to comment.