AGU citation styles: in text citations changed from square brackets to parentheses

I have been using the Global Biochemical Cycles citation style. Several months ago, when I was preparing this paper for initial submission, zotero rendered authors styles in the format [Kwon et al. 2009], with square brackets. Now, it renders them with parentheses (Kwon et al. 2009).

Between when things were working normally and now, I have switched to Zotero 5 standalone from the Zotero 4 Firefox addon (the problem could be unrelated though, I don't have a version of zotero4 to compare against). I checked with the AGU citation style and got the same behaviour. Indeed, I also get this behaviour with the Elsevier Vancouver author-date, which according to this post https://forums.zotero.org/discussion/36537/text-references-with-square-brackets should also use square brackets. I've tested this on both my windows virtual box and on a host linux system with the same behaviour. I also get this same behaviour if I go into Zotero and try to export a bibliography from item and then click citation.

Any ideas about what could be going on?

I thought I could perhaps remedy this problem for myself by modifying the .csl stile-sheet, but it looks like the stylesheet for GBC just has some links to data elsewhere on the internet, rather than containing the specifications itself. Is there a correct way to modify such citations? They seem very mysterious to me.

Happy to provide any other information as requested. Thanks!
  • AGU changed its style to APA format, i.e. regular parentheses are correct, see http://publications.agu.org/brief-guide-agu-style-grammar/#reference

    This also applies to Global Biochemical Cycles, see the red note here:
    https://gbc-submit.agu.org/cgi-bin/main.plex

    You're misreading the post you link to, which is about the regular Elsevier Vancouver, not the author-date variant. The Vancouver author-date one is correct with regular parentheses, too.

    So everything working exactly as it should. If you do want to modify a style, modify AGU (GBC just links to that).
  • Huh. Who knew. Apparently you did. Thanks for pointing that out!
  • One other thing about this format, that is new, and I think not correct. I notice that it is disambiguating author names based on first names. So if we have one paper written James Smith 2014 and another by Gerry Smith 2010, they disambiguated as

    (James Smith et al. 2015) and (Gerry Smith et al. 2010).

    I think the correct format for this journal is to never put first names, so it should be
    (Smith et al. 2015) and (Smith et al. 2010).

    Of course this is actually a problem for me due to my database being littered with ambiguous first names, and I see that this is a problem that is unlikely to be fixed. However, I don't think I should have to disambiguate all of the first names, since I don't think first names should go in the in text citations at all.
  • edited September 24, 2017
    That is correct APA style, which AGU now uses. You should leave the first names in your database as is (as they are correct to include—Zotero will include just the initial if that is sufficient to disambiguate). The one thing you would want to change would be to make all instances of the same person’s name consistent (always spelled out, rather than initials)—Zotero can’t tell that C. Smith and Charles Smith are the same person.
  • Thank you for the update on AGU's switch to the APA style. I could not understand why all Journal of Geophysical Research citations were different now, given that the timestamp on the citation style in the Zotero was 2016-05-24 03:21:39 for all JGR journals.

    The timestamp for all Journal of Geophysical Research journals and other AGU journals such as Global Biogeochemical Cycles should probably be changed in the Zotero style repository to avoid confusion. The change only occurred recently for AGU journals. (I see that the timestamp for the American Geophysical Union citation style in the repository is recent, i.e., 2017-09-05 01:06:39.)

    Is there any usefulness to adding a "AGU (deprecated)" citation style too, for those migrating from one style to the other?
  • The old style is no longer used, so there is no reason to keep it in the CSL repository in any form.
  • Agreed, but the timestamps do need to be updated to when the styles were last changed.
  • edited September 27, 2017
    The specific journal styles are just “dependent” styles that point to the master AGU style. They haven’t actually changed. The Timestamps are used to trigger updates from the repository. They are really not intended to be read and used by people.

    (In the six years I’ve been volunteering for CSL, I’ve never seen someone mention looking at a time stamp before.)
  • edited September 28, 2017
    (Please forgive my wall of text. No reply necessary.)

    I understand, but maybe the significance of time stamp is something to think about.

    I assumed that the CSL style for the JGR paper that I was preparing was broken, because the citation style in my document changed from one minute to the next. I went to update my JGR CSL from the Zotero style repository, and saw that the time stamp in the repository was from 2016, so I could not know that the style had actually changed recently.

    After downloading the CSL, I saw it was still "broken", and assumed my Zotero installation was to blame. In reality, everything was working fine, as I found out by coming to the forum and seeing that the AGU "independent" style had recently changed, and that all "dependent" styles hadn't technically changed, but that "dependent" styles simply link to the parent "independent" style. I did not know about the existence "dependent" and "independent" styles and how they work, before reading the CSL documentation and your explanation. (In my mind, all styles were self-contained, and "independent".)

    tl;dr: I had to come to the forums, and read the CSL documentation, to find out that Zotero wasn't broken, when a time stamp to show when the (parent) style (of a dependent style) was updated would have sufficed.
  • edited September 28, 2017
    I understand, but also think it's relevant that you are the first person I've ever seen look at the timestamps in a CSL file, so I'm not sure that changing the current automated timestamp behavior would do much to reduce any user confusion in a broad way.
  • I'm pretty sure we had a few people before who were confused by the fact that the timestamp of dependents hadn't updated when the independent-parent changed.

    The Zotero Style Repository already changes the timestamps in the styles (based on their file modification date). It probably wouldn't be much work to change it for dependent styles to take either the file modification date of the dependent or its parent, whatever is more recent.
  • Does anyone know the previous style used by AGU?

    Regardless of timestamps, it may be good to maintain an AGU (deprecated) style for people like myself who are currently preparing an article for submission to an AGU journal. The new style has messed up all my citations, owing to the first-name disambiguation issue discussed above. I would prefer not to have to go through every single entry in my Zotero database and manually disambiguate the first names, especially working under a huge deadline to submit this article.

    Similarly, I have added modifiers in many locations to the citations, for example:

    "This is a scientific fact (c.f. Smith et. al., 2014)."

    If I add a new citation to my article, Zotero automatically updates all my previous citations to match the new AGU style, and removes the custom modifications (i.e. c.f.). Is there a solution to this problem I can use to work around it for now?

    Thank you
  • Update to the above. I have resorted to disambiguating all the first names in my Zotero database. Even so, the in-text citations are still confused and are spelling out first names in some places but not others. I have manually fixed all such citations, but still seems confused. Any ideas for why this may be happening would be appreciated! Thanks.
  • The old style can be downloaded from here:
    https://raw.githubusercontent.com/citation-style-language/styles/9a371e3437dc85e6bce08a1e46458ec5cf6cb952/american-geophysical-union.csl
    you need to change the filename as well as the self link and id to prevent Zotero from overwriting it again (use any text editor).

    We won't keep a deprecated version active on the repository, no, that creates too much of a mess.

    Additions like cf. etc. should be added using the prefix field, see https://www.zotero.org/support/word_processor_plugin_usage
    that way style changes carry them over. Similarly, manually removing things in citations you don't like messes with style updates, so I'd advise against it. If you need to do it, make it the last thing you do when editing and after removing Zotero field codes.

    Hard to say what the problem with the current style is. Could be that you missed some items in Zotero or could be that some references are stored in the document, unlinked from Zotero (in which case you can't update the names -- the only way would be to delete & re-insert those citations).
  • adamsmith -

    Thank you very much for these helpful resources. I agree it is very strange about the first names issue I am still having. I was able to resolve all the disambiguations by manually editing the author names in my Zotero database, save for one author that is still problematic. I have four separate articles by this author in my database, and I cite three of them in my article. I have manually edited his name in my Zotero database so it is identical in all articles. I even deleted and re-imported the citations into Zotero, and then deleted and re-added them to my Word document. Still, it is writing his full name in one location, and his initialed name in another. I have copied and pasted his name so there are no spelling issues, and made sure there aren't any blank spaces before or after. I even checked every single article in my Zotero database that he is a co-author on. Still a problem.

    I suspect it might be related to the citation style of the source journal. One of his articles is from a Copernicus style journal, the others are Wiley and use APA. When I re-imported the Copernicus article, the citation uses his initials in the author tab, whereas the Wiley articles spell his name out.

    Other than that, thank you for the prefix suggestion, that is very helpful.
  • You should edit the names in Zotero to be identical, even if they import differently from different databases. Zotero has no way of knowing that the full name and initials versions of the author are the same, and so it will treat them as different authors (leading to the disambiguation behavior you are seeing).
  • bwiernik -

    As noted, I have edited them to be identical, and they are still being treated differently by Zotero. I noted that I tried deleting and re-importing (and then editing the names to be identical) as an additional step to see if that helped, but it didn't.

    Just to clarify, I have edited the name to be identical in every publication I have in my database, including where the author is included as a co-author (even though I don't think that should matter).
  • Okay, I was confused by your comments about Wiley/Index Medicis. What is likely the case is that the item showing initials is not connected to your library. Try to delete that citation and re-insert it from your Zotero library.
Sign In or Register to comment.