sort order versus repeated authors omitted

Dear all,

i have the following problem:

I would like to have the citation in the sort order: year+author and additionally have repeated authors omitted. This, however, shall have the priority, so that a citation in the text can be read as: [e.g., Li and Keller, 1999; Friedrich et al., 2004; Miller et al., 2003; 2005].

At the moment it does this:

[e.g., Li and Keller, 1999; Miller et al., 2003; Friedrich et al., 2004; Miller et al., 2005].

Is there any way of telling the output style to do this?

Thanks in advance!!


From your example it seems that the sort order needs to: 1) account for references involving more than 2 authors, while 2) sort on the first author for references involving multiple authors. If this is the case, using “Author” in the sort order of Year+Author won’t accomplish this, resulting in the citation order as: Li and Keller, 1999; Miller et al., 2003; Friedrich et al., 2004; Miller et al., 2005.

A workaround (which was hastily tested using the 4 examples) might be to create a custom field for each reference type (e.g., “3PlusAuthors”) to flag references with multiple authors and enable sorting them by the first author. Then for only references which have 3 or more authors, open the reference and in the 3PlusAuthors field, enter the name of the first author. So for example, the Friedrich et al reference’s 3PlusAuthors field will have: Friedrich.  Make similar changes to other references having 3 or more authors.

Then change the citation sort order to: Year + Custom #  (the number should correspond to the 3PlusAuthors placement in the template.

As mentioned earlier, testing was performed using 4 examples but suggest additional references to fully vet the custom field and the resulting sort order. 

Note: A variation of this solution using the custom field “2ndAuthor” to permit sorting on the second author was posted at

I did all you suggested, but when the sort order for citations is year + custom than I still get (Li and Keller, 1999, Miller et al. 2003, Friedrich et al. 2004, Miller et al., 2005)

because it is first sorted by year, and the Friedrich citation remains in the middle.

However, I wrote for sort order: Custom 3 + Year, but then I get:

(Li and Keller, 1999; Friedrich et al., 2004; Miller et al., 2003; Miller et al., 2005) and it does not omit the second Miller citation as it should. 


I just saw, that it never omits the second citation when two from the same are used (Miller et al., 2003; Miller et al., 2005) instead I would like to have: (Miller et al., 2003, 2005)

But now that both Miller citations are in a row, I can tell him to exclude the author name from miller et al. 2005, so I get what I want, but not automatically…

Now I have the problem when I have citations like these:

[Clarke and Jenkyns, 1999; Huber et al., 1995; Huber et al., 2002]

where I put “Huber” in the 3plusField, but the Huber citations should come before Clarke, because the first is earlier.

It is a bit frustrating…

I have to be honest here. 

Why does this matter so much to you?  Spending so much time on this when you should be concerned about the content of your writing, just doesn’t make sense to me (speaking as a “boss” right now).  Not to mention that these are very strange aesthetics!  If this is for a publication, rest assured that the publisher will have a mechanism to make your references sorted in the house style.  That is what editors are paid for. 

Oh, are you sure, that this is the editors work? My supervisor says differently, he says the citations need to be in correct order etc. He does everything by hand and is not using endnote, but I thought it should be possible here and even easier…

Do you have author instructions that dictate these requirements?  Publishers need to get a life. 

In my 35 years of publishing (in scientific journals), I have NEVER had a paper rejected because the citations were in a slightly inconsistent manner and yes, on occasion, the proofs came back with some rearrangements. 

If it is important to you (or to your supervisor), when they do accept it, make these perverse um, detailed changes when you send in the final version and be done with it.  I haven’t heard this kind of discussion since my son graduated from high school (where teachers appear to love imposing their own take on MLA, APA and the like and didn’t like "Endnote"s rendering.  My son won those arguments, for the most part, by the way.). 

Just my own two sense cents,  sorry for the rant. 

To clarify my earlier posting, refer to Image 1 for the steps.  Pls. note a correction to the the citation sort order which should be: Custom 3 + Year.  My apologies for inverting the order.

The end result is shown in image2.  If discrepancies continue, please identify the full author list for citations involving 3 or more authors as something may be occurring which cannot be discerned by seeing the first author alone (e.g., Frederich et al).

And ditto Leanne’s comment about publishers needing to “get a life” (and supervisors). 

Image2_In-text Citation Result in MS Word.gif


Your prior postings identified 3 citations in their preferred format:

     (Li and Keller, 1999; Friedrich et al., 2004; Miller et al., 2003; 2005)

     (Clarke and Jenkyns, 1999; Huber et al.,1995; Huber et al., 2002)

     (Miller et al., 2003, 2005)

After some experimenting it seems the 3 citation formats can be generated, however, it’s hard to say whether the initial set-up to achieve these citations will hold up when tested against with other variants which haven’t been identified.  The key difficulty occurs with “splitting” references by a repeated author such as:  (Clarke and Jenkyns, 1999; Huber et al.,1995; Huber et al., 2002) - which also requires an alphabetical sort to place Clark and Jenkyns first.  It wouldn’t be a problem if all repeated authors are to be treated this way but the first citation shows the condiion of aggregating repeated authors: (Li and Keller, 1999; Friedrich et al., 2004; Miller et al., 2003; 2005). So the two  

Again, it’s possible to replicate the 3 citations listed above but due to the setup, problems will occur if Huber (1995, 2002) needs to be cited elsewhere in aggregate form.  Constructing a solution (or knowing when there’s no solution) requires identifying all the possible citation format conditions and just resolving these 3 citations has gotten very complicated. I’m doubtful that the setup for generating the 3 citations will accommodate other conditions which may need to be identified and addressed .

Given the uncertainty and amount of work involved in trying to adjust the citations - and if your supervisor still insists on this format - suggest you make a copy of your document, convert  it to text to remove the EndNote field codes, then manually make the necessary changes.