Since I've been using Mendeley to manage and extract metadata from my PDFs, I've come to notice that authors are sometimes denoted with all their initials and sometimes with just one. I assume this has to do with different journal's rules on denoting author's names. However, this makes a mess in my author list in Mendeley, as the same person may be known under two or even more names (e.g.: "Last, F."; "Last, First"; "Last, First M,"; "Last, First Middle").

If, for completeness' and tidyness' sake, I edit the author fields to always have the most complete name of the author available to me regardless the journal the author published in, I inevitably will change the way I cite them. For example, an article from a certain journal only uses the first initial:

Last, F. Article Title, Journal name, x:y (YYYY), pp. xxx-xxx

After I have "updated" my authors, it is possible my citation includes also the second initial, i.e.:

Last, F.M., Article Title, Journal name, x:y (YYYY), pp. xxx-xxx

So my question is: am I incorrectly citing the article when I include more complete information about the author's name than is given by the journal article itself?

5 Answers 5


I can see your problem but I am afraid you will have to live with the mess. The reason is that a publication must be referenced exactly the way it is published. The exception is that it is permissible to abbreviate first names to initials to adhere to the standard of the publication in which you intend to publish. It is, however, not allowed to remove middle names (initials) if they are part of the original publication, nor to add initials if they were not part of the original. I know many authors who have published under one and two initials randomly, but there is not much one can do about it.

Now, if you do it in your reference program, there is nothing wrong with that, but you need to make sure your changes do not migrate into published work that you write.

Just an example, Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Manuscript Preparation and Submission: Preparing a Manuscript for Submission to a Biomedical Journal states:

Some but not all journals check the accuracy of all reference citations; thus, citation errors sometimes appear in the published version of articles. To minimize such errors, references should be verified using either an electronic bibliographic source, such as PubMed or print copies from original sources. [The text then goes on to discuss retracted articles.]

  • 22
    "a publication must be referenced exactly the way it is published": why? Jul 3, 2013 at 19:38
  • 14
    But the exact source is typically clear even if you did not have any author's names at all in the bibliography. For example, journal name + journal volume + page numbers are typically sufficient to identify the article that you are citing. With the exact title of the article and the correct last names of the authors, there is very little room for any kind of confusion or ambiguity left. I cannot really see how it would confuse anyone if you used the full first name instead of some kind of abbreviated version of the first name, no matter what was written in the original article. Jul 3, 2013 at 19:50
  • 18
    But citations serve multiple purposes. Not only are you telling the exact source for a piece of information that you are using; you are also using citations to give credit to those who made the discovery. And I think the use of, e.g., full first names is a much more effective way of giving credit to the right people. Jul 3, 2013 at 19:55
  • 18
    It is, however, not allowed... That sounds like there is one authority who's (de facto) putting down these rules. Is there something like that? Otherwise I would guess these rules are a bit more squishy and up to pragmatism or even personal style. Jul 14, 2014 at 8:19
  • 5
    You can run into problems this way. There are two common transliterations of the Russian name Делоне, Delone and Delaunay. Suppose you want to cite two translated articles, which used different transliterations, and one untranslated article. What do you do? Jul 17, 2014 at 13:21

I disagree with the statement of "a publication must be referenced exactly the way it is published."

From Knuth's Mathematical Writing

In his bibliography Knuth has tried to keep his citations true to the original sources. The bibliography contains mathematical formulas, full name spellings (even alternative spellings when common), and completely spelled-out source journal names. (This last may be unusual enough that some members of a field may be surprised to see the full journal name written out, but it’s a big help to novices who want to find it in the library.)

This means that you can change reference, give full name of authors, give full name of journals. It may be unusual but it can be done and it is DONE. It is done by one of the most respected authors in the academy.

To answer your question:

So my question is: am I incorrectly citing the article when I include more complete information about the author's name than is given by the journal article itself?

When you include more complete information, you are not doing anything wrong. Actually according to Knuth, you are doing very good job.

Therefore if your target journal does not have any rule about same reference rule, you should be perfectly alright.

  • 13
    +1. I am actually quite religious about hunting down authors' full names and using them in my bibliographies, even when they publish using only their initials. (Of course, this requires care; some people's given names are initials.)
    – JeffE
    Jul 17, 2014 at 2:08
  • 2
    In recent years, I've tried to give authors' full names, or at least the first and last names (middle names can be too hard to find). As far as I can remember, one journal copy-edited the first names down to initials; no other journal made any changes or asked me to make changes. Jul 17, 2014 at 4:09
  • @AndreasBlass, but your personal bibliography database retains full data.
    – vonbrand
    Dec 29, 2015 at 18:55
  • 1
    Your quote doesn't seem to me to match the rest of your answer. It sounds like Knuth is saying "use the full version from the original source rather than an abbreviated version". After all, it is prefixed by being true to the original source. If the original was in "Journal of Awesome Results" by "Donald E. Knuth", don't cite it as "J. Awes. Res." by "D. E. Knuth". But also don't cite it by "Donald Ervin Knuth" unless that appears on the paper. I had a look at some bibliographies in Knuth's solo papers, and they are consistent with my understanding. Sep 8, 2023 at 12:45

You should not add an initial that was not present in the original publication or expand initials into full names if the author used initials. One reason is the one Peter Jansson gives, namely that the bibliography should reflect what was actually published. However, there's another reason I find even more compelling: respect for the author's choice of name.

Editing another person's name is presumptuous. Like many people, I have deliberately chosen the version of my name I use in papers, and I care about this aspect of my professional identity. What my passport says or what I'm called in person are irrelevant, and I would not be happy to have citations edited to use what someone else thinks is a better (or more formal, or more complete) version of my name than the one I chose to publish under. If my paper says Alice P. Liddell, then nobody has any business deciding I should be called Alice Pleasance Liddell instead.

In particular, your desire for completeness or consistency should not outweigh the author's autonomy to choose their own professional name, and you should not overrule a deliberate choice. Of course, some people aren't picky about what they're called, and they may publish using several random variants of their name. It can be tempting to standardize the name for the sake of consistency, but you should avoid doing so unless you know the author wouldn't mind. If you don't know the author personally, it's hard to distinguish between an author who doesn't care and one whose preferences have changed over time, and it's not reasonable to rely on your own guess as to how to handle this. (For example, if someone adds an initial due to marriage, they might be displeased to see it retroactively applied as if it had always been their name.)

There can also be political aspects of naming. For example, some women publish under initials to avoid drawing attention to their gender. I'm not convinced this makes a difference, but who am I to unilaterally undo it by replacing the initials with identifiably female names?

To summarize, names can be a sensitive subject. Every paper comes with the names chosen by the authors at the time of publication. If you're going to modify these names, you'd better have a good reason.

  • 18
    respect for the author's choice of name — This is a fair point. However, in many cases, the choice to use initials instead of full names is not made by the author, but by the journal. I've had to fight with copy editors to use my full first name in my own papers.
    – JeffE
    Jul 17, 2014 at 19:02
  • 6
    This holds water as far as introducing new spellings goes, but one of the drives of the OP is standardising spellings within a database, which amounts to choosing among variants which the author has publicly sanctioned. (On the other hand, there are those times where you just have to budge on spellings or whatever, and having others unknowingly seize on the variant you'd rather not use isn't fantastic.)
    – E.P.
    Aug 27, 2014 at 22:59
  • 1
    This can cause problems, or at least oddities, with multiple citations from a single author if two or three variants show up. To me, if I see nested entries for a single author, I don't necessarily presume the author published with the exact same name in each. But if I see two or three similarly, but not identically, spelled names with unique single author entries, I presume that they are indeed three unique people, and that's not a good thing. Dec 29, 2015 at 16:48

I would be very careful altering the form of the author names given in a publication. As others have stated, it is easier to trace citations if they match those given by the journal. It is also concern for consistency, especially for common names, it would be laborious to authenticate that these are indeed the same authors and not those with the same initials working in related fields across your entire reference library or bibliography.

It is also difficult to deduce "journal rules" from looking at a few articles as many cultures also do not have the custom of middles names. Although I have heard of Japanese authors also taking advantage of this to use their nickname as a middle initial to distinguish themselves from an existing author with a similar name (such as Watal "Metal" Iwasaki). Similarly, some people have geniune cultural reasons for changing their legal or pen name.

Another consideration is that authors may be very deliberate reasons for including their name or initials they way they have in a publication. Some people (including myself) are known by their middle name but publish under their full legal name (e.g., S. Thomas Kelly). These do occur in prominent papers, such as J. Craig Venter who sequenced one of the first genomes. Removing the middle initial may be problematic when attributing such people as it is important for their identity and to show that their work has been cited later.

If it is really relevant to your argument to link publications, I'd recommend mentioning that the articles are from the same author or research group when citing in text. It's far easier to remain consistent with citations exported from a reference manager and will demonstrate your critical engagement with the literature if you compare articles in this way.

  • 2
    "it would be laborious to authenticate that these are indeed the same authors and not those with the same initials working in related fields" - it can get worse: I know of at least two such author pairs who even work in the same department, researching the same topics. In one case, they only differ by a middle initial - one goes by "First Last", the other by "First M. Last". That makes it very difficult for someone not familiar with them to realize they are not the same person. Mar 23, 2017 at 13:17
  • 1
    Conversely, there are also cases where one person publishes inconsistently under several different names. I didn't say it would be easy.
    – Tom Kelly
    Mar 24, 2017 at 3:15

The problem is that authors may have legitimate reasons for wanting to use only initials. Maybe because of a clerical error, the same author is John Smith according to his birth certificate and Jon Smith, without the H, on his naturalization certificate. Maybe his name was misprinted as Jonas Smith on a widely-cited previous paper. Or maybe he intends to eventually be Jane Smith.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .