Natbib problem with same year citations (Author, 2001a,b,?)
Jerry Fountain
I'm having a problem with Natbib (I think) when using lists where
the "same" authors are listed multiple times in the same year. I'm
using the latest Natbib (just pulled the latest to check) on a PC
with MikTeX, and using a chicago style (bib file with references).
The chicago file is older (1992) with the author name of Glenn
Paulley in case this helps. I'm a bit of a Latex novice (I can
use it, but not program it) so bear with me.
Everything works fine normally. References work well, no major
complaints out of BibTeX when compiling the *.bib file. I've got
a few cases through where a first author (with coauthors) are
listed multiple times in the same year. Since there are
coauthors, this results in an "et al." type listing. When all the
coauthors are the same the reference works fine (Jones et al.
1999a,b). But, if the coauthors are NOT the same, then I get the
following in the dvi output (Jones et al. 1999,?).
Anyone care to point me in the right direction to fix this? I'd
really appreciate it. With Deja gone "Googlie" I haven't been
able to find any references to this problem. I think this is
controlled by natbib, but as I said, I can use the features, but
sometimes the *.sty and *.bst hierarchies are a bit fuzzy :)
Oh, btw, if I wanted the "et al." listings to be in italics, where
would I change this?
Thanks in advance!
Jerry
--
Jerry Fountain | Laboratory for Fluid Mechanics, Chaos, and Mixing
g...@chem-eng.nwu.edu | Northwestern University
(847) 491-3555 (Office) | Department of Chemical Engineering
(847) 491-3728 (FAX) | 2145 Sheridan Road, Evanston, IL 60208
Una Smith
>
>I'm having a problem with Natbib (I think) when using lists where
>the "same" authors are listed multiple times in the same year. I'm
>using the latest Natbib (just pulled the latest to check) on a PC
>with MikTeX, and using a chicago style (bib file with references).
Use makebst.tex in the custom-bib package to make a .bst for the
style you want that is is also compatible with natbib.
--
Una Smith una....@yale.edu
Department of Ecology and Evolutionary Biology
Yale University
Robert Schlicht
>
> I'm having a problem with Natbib (I think) when using lists where
> the "same" authors are listed multiple times in the same year. I'm
> using the latest Natbib (just pulled the latest to check) on a PC
> with MikTeX, and using a chicago style (bib file with references).
> The chicago file is older (1992) with the author name of Glenn
> Paulley in case this helps. I'm a bit of a Latex novice (I can
> use it, but not program it) so bear with me.
>
> Everything works fine normally. References work well, no major
> complaints out of BibTeX when compiling the *.bib file. I've got
> a few cases through where a first author (with coauthors) are
> listed multiple times in the same year. Since there are
> coauthors, this results in an "et al." type listing. When all the
> coauthors are the same the reference works fine (Jones et al.
> 1999a,b). But, if the coauthors are NOT the same, then I get the
> following in the dvi output (Jones et al. 1999,?).
>
You can either swap citations; that will give (Jones et al. 1999,1999b),
or you could simply cite them seperately, which will give (Jones et al.
1999, Jones et al. 1999b). The latter is probably better as it also
reflects the fact that Jones et al. doesn't mean the same in the two
citations.
You could try other bib styles, but I think most (at least from the
standard ones) won't even recognize that the two labels refer to
different groups of authors, if they abbreviate them with "et al.".)
> Anyone care to point me in the right direction to fix this? I'd
> really appreciate it. With Deja gone "Googlie" I haven't been
> able to find any references to this problem. I think this is
> controlled by natbib, but as I said, I can use the features, but
> sometimes the *.sty and *.bst hierarchies are a bit fuzzy :)
>
natbib is innocent. From the labels that chicago.bst produces, it simply
cannot tell that they are supposed to be different. And so it puts a "?"
in your reference as it expects ambiguous references to be labelled with
a letter.
> Oh, btw, if I wanted the "et al." listings to be in italics, where
> would I change this?
>
For labels edit the function {format.lab.names} in chicago.bst by
changing the line
{ s #1 "{vv~}{ll}" format.name$ " et~al." * }
to
{ s #1 "{vv~}{ll}" format.name$ " {\em et~al.}" * }
For changing the appearance in the bibliography you need to change the
function {format.names} accordingly. (This only applies to when you gave
something like "Jones, Stephens and others" in your bib entry.)
> Thanks in advance!
>
> Jerry
Hope this helped,
Regards,
Robert.
--
Robert Schlicht
<---------------------------------
schl...@informatik.hu-berlin.de
--------------------------------->
Jerry Fountain
Robert Schlicht <schl...@informatik.hu-berlin.de> wrote:
>Jerry Fountain wrote:
>>
>> I'm having a problem with Natbib (I think) when using lists where
>> the "same" authors are listed multiple times in the same year. I'm
>> using the latest Natbib (just pulled the latest to check) on a PC
>> with MikTeX, and using a chicago style (bib file with references).
>> The chicago file is older (1992) with the author name of Glenn
>> Paulley in case this helps. I'm a bit of a Latex novice (I can
>> use it, but not program it) so bear with me.
>>
>> Everything works fine normally. References work well, no major
>> complaints out of BibTeX when compiling the *.bib file. I've got
>> a few cases through where a first author (with coauthors) are
>> listed multiple times in the same year. Since there are
>> coauthors, this results in an "et al." type listing. When all the
>> coauthors are the same the reference works fine (Jones et al.
>> 1999a,b). But, if the coauthors are NOT the same, then I get the
>> following in the dvi output (Jones et al. 1999,?).
>>
>
>You can either swap citations; that will give (Jones et al. 1999,1999b),
>or you could simply cite them seperately, which will give (Jones et al.
>1999, Jones et al. 1999b). The latter is probably better as it also
>reflects the fact that Jones et al. doesn't mean the same in the two
>citations.
I tried swapping them but that doesn't work. In one case I can rearrange
the order of references (there are more than the two) to put one in between.
But it looks like in that case I'll have to explicitly place an 'a' and 'b'
in the year field. Citing them separately is problematic due to the
placement of punctuation (if it was a \citep for example, you would end up
with (Jones et al. 1999) (Jones et al. 1999).
>You could try other bib styles, but I think most (at least from the
>standard ones) won't even recognize that the two labels refer to
>different groups of authors, if they abbreviate them with "et al.".)
In the bib file, the authors are fully listed. I guess from your
description further down that the chicago.bst works with BibTex to read the
cited list from LaTex and then spits back a formatted list based on that.
When does the full list -> et al change occur? Back in LaTex it would seem.
So something there is getting confused. Or am I still missing something?
>> Anyone care to point me in the right direction to fix this? I'd
>> really appreciate it. With Deja gone "Googlie" I haven't been
>> able to find any references to this problem. I think this is
>> controlled by natbib, but as I said, I can use the features, but
>> sometimes the *.sty and *.bst hierarchies are a bit fuzzy :)
>>
>
>natbib is innocent. From the labels that chicago.bst produces, it simply
>cannot tell that they are supposed to be different. And so it puts a "?"
>in your reference as it expects ambiguous references to be labelled with
>a letter.
>
>> Oh, btw, if I wanted the "et al." listings to be in italics, where
>> would I change this?
>>
>
>For labels edit the function {format.lab.names} in chicago.bst by
>changing the line
>
> { s #1 "{vv~}{ll}" format.name$ " et~al." * }
>to
> { s #1 "{vv~}{ll}" format.name$ " {\em et~al.}" * }
>
>For changing the appearance in the bibliography you need to change the
>function {format.names} accordingly. (This only applies to when you gave
>something like "Jones, Stephens and others" in your bib entry.)
Thanks for the help. I'll only need to change how it is formatted for the
in-text citations since the bibliography will have everyone spelled out.
You've given me something to check at least for the in-text citations. I'm
somewhat surprised that this hasn't been addressed (or fixed) long before
now. I would assume that having two papers from the same first author in a
given year to be common enough for this to have come up many times before.
I've come to expect (in my limited use of LaTex) that everything works
perfectly. :)
Jerry Fountain
Jerry Fountain <g...@jules.chem-eng.nwu.edu> wrote:
>In article <3AC7BF26...@informatik.hu-berlin.de>,
>Robert Schlicht <schl...@informatik.hu-berlin.de> wrote:
>>
>>You can either swap citations; that will give (Jones et al. 1999,1999b),
>>or you could simply cite them seperately, which will give (Jones et al.
>>1999, Jones et al. 1999b). The latter is probably better as it also
>>reflects the fact that Jones et al. doesn't mean the same in the two
>>citations.
>
>I tried swapping them but that doesn't work. In one case I can rearrange
>the order of references (there are more than the two) to put one in between.
>But it looks like in that case I'll have to explicitly place an 'a' and 'b'
>in the year field. Citing them separately is problematic due to the
>placement of punctuation (if it was a \citep for example, you would end up
>with (Jones et al. 1999) (Jones et al. 1999).
Folowing up myself. I tried explicitly putting in year fields of 1999a and
1999b. Doesn't work. The thing seems to take the trailing 4 characters so
the years were abbreviated to 999a (actually, the real one was 1981, so I
got 981a and 981b).
Back to the drawing board...
Robert Schlicht
>
> In article <3AC7BF26...@informatik.hu-berlin.de>,
> Robert Schlicht <schl...@informatik.hu-berlin.de> wrote:
> >Jerry Fountain wrote:
[...]
> >> When all the
> >> coauthors are the same the reference works fine (Jones et al.
> >> 1999a,b). But, if the coauthors are NOT the same, then I get the
> >> following in the dvi output (Jones et al. 1999,?).
> >>
> >
> >You can either swap citations; that will give (Jones et al. 1999,1999b),
> >or you could simply cite them seperately, which will give (Jones et al.
> >1999, Jones et al. 1999b). The latter is probably better as it also
> >reflects the fact that Jones et al. doesn't mean the same in the two
> >citations.
>
> I tried swapping them but that doesn't work. In one case I can rearrange
> the order of references (there are more than the two) to put one in between.
> But it looks like in that case I'll have to explicitly place an 'a' and 'b'
> in the year field.
As you found out yourself, this doesn't work. But see below.
> Citing them separately is problematic due to the
> placement of punctuation (if it was a \citep for example, you would end up
> with (Jones et al. 1999) (Jones et al. 1999).
>
What about (\citet{jon99}, \citet{jon99a}) ?
> >You could try other bib styles, but I think most (at least from the
> >standard ones) won't even recognize that the two labels refer to
> >different groups of authors, if they abbreviate them with "et al.".)
>
> In the bib file, the authors are fully listed. I guess from your
> description further down that the chicago.bst works with BibTex to read the
> cited list from LaTex and then spits back a formatted list based on that.
> When does the full list -> et al change occur? Back in LaTex it would seem.
> So something there is getting confused. Or am I still missing something?
>
The confusion does not happen in BibTeX. The output that BibTeX produces
together with chicago.bst is correct insofar as it considers, say,
"Jones, Smith and Stephens" as distinct from "Jones, Smith and
Michaels", so that it doesn't label the years with letters. This
complies with academic standards, I suppose (but correct me if I'm
wrong).
The confusion comes in because chicago.bst wasn't actually made for use
with natbib. If you only used chicago.bst (together with chicago.sty)
everything would be as expected, only that citations are not aggregated
like natbib does it. That is, with \shortcite{jon99, jon99a} you would
get: (Jones et al. 1999, Jones et al. 1999a), which again is confusing
of course, especially if Jones had written more than two papers in one
year together with some students who did the main work but whose names
are replaced by "et al.".
The problem is that chicago.bst calculates both an abbreviated and a
long label, but natbib doesn't care about the long label if you want a
short label and thence gets the latter wrong. Natbib considers every
label that consists of the same string ("Jones et al.") to refer to the
same authors so that it insists on letters after the year which
chicago.bst doesn't give because it has a different philosophy. Phew!
OK. But lets get to a solution. I have four possibilities for you:
-- You could use \usepackage{natbib}, \bibliographystyle{plainnat} and
put \citationstyle{chicago} before \begin{document}. You will get
letters after the year for every paper of Jones, regardless whether his
coauthors are the same or not. You then only have to hope that it comes
close to your desired bibliography style.
-- If it doesn't, and you want to stick to chicago.bst, you could add
letters to your .bbl-file (not .bib). The bbl-file is BibTeX's output
and LaTeX's input, so it will be overwritten every time you bibtex your
paper, and there's not much sense in editing it just before the final
LaTeX run. You have to specify the letters both in the label and in the
bibliography entry of course.
-- As alread said, you could use \usepackage{chicago} and
\bibliographystyle{chicago}, which would give you no letters for
different groups of authors.
-- Or you could also refer to (and revere): et altera.
[...]
>
> You've given me something to check at least for the in-text citations. I'm
> somewhat surprised that this hasn't been addressed (or fixed) long before
> now. I would assume that having two papers from the same first author in a
> given year to be common enough for this to have come up many times before.
> I've come to expect (in my limited use of LaTex) that everything works
> perfectly. :)
>
Regards,
Robert Schlicht
>
> Jerry Fountain wrote:
> >
> > In article <3AC7BF26...@informatik.hu-berlin.de>,
> > Robert Schlicht <schl...@informatik.hu-berlin.de> wrote:
> > >Jerry Fountain wrote:
>
> [...]
>
[...]
> OK. But lets get to a solution. I have four possibilities for you:
> -- You could use \usepackage{natbib}, \bibliographystyle{plainnat} and
> put \citationstyle{chicago} before \begin{document}. You will get
^
|
this should read \citestyle{chicago}. Sorry.
Robert.
--
Robert Schlicht
<---------------------------------
schl...@informatik.hu-berlin.de
r_sch...@yahoo.com
--------------------------------->