6

It is commonly taught (e.g. in the "Guide to LaTeX" (4e), p. 235) that names with a "junior" part should be given in the form author = "Surname, Junior, Given" in a BibTeX entry.

How should "junior" parts appear in bibliographies with inverted (surname-first) name order? The widely used style guide "A manual for writers of research papers, theses, and dissertations" (Kate L. Turabian; 8e (2013); 978-0-226-81638-8) has an example on page 152:

Gates, Henry Louis, Jr. America behind the Color Line: Dialogues with African Americans. New York: Warner Books, 2004.

While BibTeX's plain style produces no inverted name order

[1] Given Surname, Jr. Title. Publisher, 1000.

all of the few styles with an inverted name order produce a name-internal "junior" part:

authordate1:

Surname, Jr, Given. 1000. Title. Publisher.

apa:

Surname, Jr, G. (1000). Title. Publisher.

chicago:

Surname, Jr, G. (1000). Title. Publisher.

kluwer:

Surname, Jr, G.: 1000, Title, Publisher.

This can't be right. Turabian is just one of many prescriptive style guides, but there is also a good semantic reason to always have name-final placement of the "junior" part, as in "Surname, Given, Jr <rest of bibliographic entry>": the "junior" part modifies the entire name, not just a part of it. So it shouldn't be in the middle of a surname-given cluster, whatever the order in which it is presented. Or maybe it can be said to modify only the given name. But in this case, too, there is no good reason to have the "junior" part in the middle: The hierarchical structure of the name is something like "surname > given > junior" (which I suppose can be interpreted as "[surname > given] > junior" or "surname > [given > junior]"), and therefore any ordering that messes with this is odd.

Here is the code used to generate the above results:

\documentclass{article}
\usepackage{natbib}
  % this line seems to be necessary for some styles for error-free compilation;
  % I welcome pointers to explanations of this unexpected dependence


\begin{document}

Text. \cite{testentry}.

\bibliographystyle{kluwer} % choose from: plain, authordate1, apa, chicago, kluwer, ...
\bibliography{testbibliography}

\end{document}

testbibliography.bib:

@book(testentry,
  title = "Title",
  author = "Surname, Jr, Given",
  publisher = "Publisher",
  year = 1000,
)

(Different point: On page 333, Turabian's manual states that "[i]n frequent references to a father and a son, shortened versions may be used (Holmes Sr.), but only after the full name has been presented". The styles I tried above don't attempt this either. You can see this by adding another bibliographic entry testentry2 with Sr instead of Jr and then referring to it in the main document's body.)

Lover of Structure
  • 22,025
  • 1
    In my experience, I see the 'jr.-part' come at the end of the name. Luckily -- especially since I never noticed this detail before -- I stopped using plain BibTeX-based styles a long time ago; and biblatex-chicago, e.g., gets it right. (Anyway, in my opinion, the most important point about bibliographies is consistent presentation of bibliographical data so that others can find it again at a later date. Virtually everything else is personal preference. Thus, while your reasoning makes sense to me, there's an old saying that starts 'de gustibus...'.) – jon Jun 05 '14 at 04:54
  • 3
    The kluwer.bst bibliography style is a natbib style, which is why you need to load natbib. With both biblatex-apa and biblatex-chicago the name suffixes are correctly placed (after the first name/initials). "Correctly" = in accordance with the APA and Chicago Manual of Style, respectively. – Alan Munn Jun 05 '14 at 05:03
  • 1
    Pardon the rant, but over my career, I've always been amazed at how editors can look you straight in the eye and assert why this or that fine point of bibliography style is the "best". I believed them the first 12 times, until (over the years) I started seeing flip-flops back to what had been prior "unacceptable" formats. I guess it's a measure of job security to \renewcommand on the bibliography styles, and I thank Knuth and others for enabling a BibTeX mechanism that can help alleviate the whimsy to which authors are subjected. And by the way, the well respected Turabian is wrong! 8^) – Steven B. Segletes Jun 05 '14 at 10:30
  • 2
    The fetishization of bibliography formatting is something that never ceases to amaze me, too. This leads to 'cargo-cult' citing practice, where it's more important to format the references than to actually cite the relevant literature itself. It's true, however, that in the humanities there is a much wider range of reference types than we see in the sciences, but that doesn't excuse the wide variety of formatting decisions. – Alan Munn Jun 05 '14 at 16:03
  • 1
    This seems to be more of a style question than anything technical. The format described by the BibTeX docs is for parsing input for BibTeX, and in that sense is correct (it is how BibTeX determines which is the Jr part). What then happens for the output is up to the person defining the style used, and that's not a technical issue. – Joseph Wright Jun 05 '14 at 18:31
  • @jon And the fact that one often uses a comma to separate the "jr" part from what's before is in my opinion consistent with (and weak evidence for) it modifying the entire name. – Lover of Structure Jun 09 '14 at 18:34
  • @AlanMunn And authordate1 also works with \usepackage{authordate1-4}. – Lover of Structure Jun 09 '14 at 18:34
  • @StevenB.Segletes I didn't believe them the first dozen times even. If it were up to me, I'd be completely doing my own thing in all respects (because I disagree with style guides and much of prescriptivism on many points), and every so often I get told off for trying. – Lover of Structure Jun 09 '14 at 18:36
  • @AlanMunn What bothers me most is that the common formatting styles don't have provisions for the things I actually want to do: indicate pseudonyms, indicate special titles of journal issues ("Special commemorative issue for ...", "Selected papers from ..."), give separate doi's for issues and articles, give version numbers for documents published online, have a unified bibliography-initial (or -final) index of journal (or other) name abbreviations, have separate fields for "form of name used in publication" and full/actual author names, omit month or indicate day of publication ... – Lover of Structure Jun 09 '14 at 18:40
  • @AlanMunn ... differently for different journals that might just be doing it differently, have an "imprint of" field and a "formerly known as" field, have links to newer web versions (with version numbers) of articles, have "organizer of conference" and "owner of journal" fields, indicate conference dates, have a format for conference presentations, have a field for "date retrieved". Not all of these are always important, and I don't actually care about consistency in formatting as much as I do about being able to put ... – Lover of Structure Jun 09 '14 at 18:44
  • @AlanMunn ... all that stuff in that I think belongs and will be helpful to a reader. And what about subentries of a bibliographic entry? Maybe they have the same author, or maybe only one or all authors of book chapters are different. And so on. But, no, I have to fiddle with the "note" field and abuse other fields and will be certain that things will break as soon as I change the bibliography style. I'm also unhappy with bizarre styles that make it hard to parse information for the human reader. – Lover of Structure Jun 09 '14 at 18:48
  • @JosephWright I didn't ask about the BibTeX file format, I meant the bibliographic styles commonly used with BibTeX. I think I have, in this case, good reason to believe that they've gotten it wrong. I'm quite descriptivist in most respects, but if there's no easy way to get things right with certain standard surname-first styles, that's a problem. – Lover of Structure Jun 09 '14 at 18:50
  • @LoverofStructure My point was that the styles supplied with BibTeX work in the way they were designed to (i.e. format.name$ does produce output it should following the requirements in .bst files): what you feel should happen in a stylistic sense is not a technical question, so would make this 'off topic'. – Joseph Wright Jun 09 '14 at 21:48
  • 2
    I too am sometimes bit by the desire to include 'extra' information in the bib (partly because I want to learn how to do it, if needed); but it is worth remembering that just because a book (say) has all kinds of extra bibliographical information that could be included, it doesn't mean it must or even should be included. A bib entry only needs to give enough non-ambiguous information that an interested reader can find it for him-/herself. Thus, consistency is not a goal in and of itself, but only an aid to that end. Of course, as you know(?), biblatex gives you the chance to include it all – jon Jun 10 '14 at 00:14
  • 2
    This question appears to be off-topic because it is about typesetting rules rather about how to achieve them in BibTeX. – yo' Oct 31 '14 at 22:54

1 Answers1

2

Both

Gates, Henry Louis, Jr.

and

Gates, Jr., Henry Louis

are entirely acceptable, in the sense that there's no possibility for confusion over what the various parts of the name are. The first form may be a bit more common than the second form, but there's obviously nothing "wrong" with either form.

Incidentally, the makebst utility, which lets you create highly customized bibliography style files, features as one of its (many) questions whether the "Junior" component should be listed before or after the "First Name" component. I think it's significant that the author of the makebst utility -- who is also the author of the natbib citation management package -- chose not to take sides as to which ordering is "correct". The choice of ordering of the components of a name is a matter of style and taste. And, as well all know (or should no), there's no arguing about taste...

Mico
  • 506,678