I've begun preparing my thesis and several key papers in my field are written by people with foreign accents in their names. While I have found a way to include this in my thesis with BibTeX. However it has come to my attention that many citations exported from online journals (as .rif, .enl, or .bib) or my previous EndNote library may not include accents in non-English names.

While I would like to attribute these appropriately it is clearly a laborious task to check for the sheer number of references included in a thesis unless strictly necessary. Therefore I have two related questions:

  1. Is it acceptable academic conduct to cite authors without correct accents when citing them in an English language thesis (or publication)?

  2. Is it common practice for reference managers or online export tools to support accents or author names (i.e., can I take it for granted that my existing library has included these characters correctly or will be necessary to check references previously imported by online databases)?

To clarify, this query concerns accents such as those found in names originating in French, Italian, Spanish, Hungarian, German, Scandinavian and Te Reo Māori languages. Such as the symbols:

ó, ò, ő, õ, ø, ö, and ō

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – eykanal
    Commented Mar 26, 2017 at 3:20
  • 1
    This question has seen a number of edits to the title, some of which subtly change the meaning of the question. Considering that the answers below more than answer the question, I'm temporarily locking this to ensure that the question stays stable while it remains in the "Hot Network Questions" list.
    – eykanal
    Commented Mar 26, 2017 at 3:23
  • 8
    Note, that there are languages where removing accents from a name turns it into a completely different name, or into a very offensive swear-word. Accents are not there as a decoration, accented letters are completely different letters in the alphabet. To give a very mild example: you wouldn't want to change the letter "K" into "H" in the name "Kate", arguing that "they look similar anyway".
    – vsz
    Commented May 18, 2017 at 10:33
  • 1
    Even more: What if the original paper is in Cyrillic characters? Do you use them in your citation? What if it is in Chinese characters?
    – GEdgar
    Commented Dec 11, 2018 at 13:28
  • 1
    @GEdgar I actually posted a follow-up question on that: academia.stackexchange.com/questions/99073/…
    – Tom Kelly
    Commented Jul 2, 2019 at 4:34

6 Answers 6


As somebody with accents in the name who does not put them on scientific papers, I would recommend going with the way they put their names on the papers.

Certain people are very particular about having the right accents, but others (including myself) consider them a nuisance and avoid them. The only way to know in the particular case is to check the paper and stick with the format on the original paper.

As far as ethics goes, I have never heard of any such policy, but there are certain people who will be your eternal enemy if you write their name incorrectly (i.e. without the right accents).

  • 31
    That is really the only reliable input in this process. If any accents got lost in the editing process, the author(s) do have the option to have it corrected at the time of publishing and or proofing. If they didn't have it corrected during proofs, they probably don't care enough about their accents anyway. Commented Mar 23, 2017 at 13:18
  • 11
    "the author(s) do have the option to have it corrected at the time of publishing and or proofing" - in theory, yes. But sometimes, the editors are simply incompetent (with respect to typesetting accented characters), and sometimes, they right out tell the authors something that amounts to "We won't do it the way you want for technical reasons." And sometimes, what is eventually published is not what the authors agreed to in the proofing version. I've encountered all of these cases, although it was only related to bibliography entries rather than my own name in my publications. Commented Mar 23, 2017 at 13:24
  • 5
    @O.R.Mapper Also, sometimes one of the authors is a "low ranking" grad student on a 15-author paper, and he simply doesn't get the chance to have his voice heard.
    – Szabolcs
    Commented Mar 23, 2017 at 13:42
  • 11
    You can also look at how it is written on their homepage, if they're alive and have one. One typically has more control over that.
    – user9646
    Commented Mar 23, 2017 at 15:59
  • 4
    Reminds me of an anecdote told by an econ professor of mine: at a seminar attended by economist Axel Leijonhufvud, trying to be courteous he'd cited a paper of his, but accidentally misspelled the name: Leijonhufvud was offended and refused to talk to him afterwards (if truth be told the professor who told the story is somewhat more eminent than the Swede).
    – PatrickT
    Commented Mar 24, 2017 at 21:16

I'm not sure how far ethics comes into this, but omitting accents is, effectively, a misspelling of the name. If you need to cite something by Schön, and you instead write "Schon", I'd regard that as analogous to writing "Schöm" or "Schöh": pretty close, but certainly a typo. If I'm reviewing a paper, I will request correction of missing accents as I would for other misspellings, but I wouldn't regard it as an ethical breach.

In practice, there's a bit more leeway for missing accents than for other misspellings, partly due to former limitations in computer typesetting, and partly due the dominance of English as a scientific language and the frequent belief among native English speakers that accents are always optional. As Burak Ulgut says, some authors won't mind missing accents, while others will care a lot -- but since you can't know which is the case for any particular author, you should keep the accents. In some cases, missing an accent may have unfortunate consequences in the author's native language: if you cite a Swedish Dr Hörberg as "Horberg", you've just turned them into "whore mountain".

For citation purposes, the correct spelling is (almost always) the one on the publication itself: if, say, Miloš Blažek chooses to publish as Milos Blazek then that's how you cite him, no matter what it says on his birth certificate.

On the technical side: unfortunately you cannot take it for granted that exported bibliographic records will handle non-ASCII characters correctly, or for that matter that they will handle anything correctly. Bibliography record formats are a mess of partially supported, poorly defined standards interconverted by buggy code, and you should always eyeball the record after importing it to catch any errors (not only accents).

As you say, it's clearly a laborious task, but so are many aspects of writing a thesis :).

  • 59
    The Scandinavian languages are particularly strong examples where "accented" letters (å, ä, ö in Swedish or æ, ø, å in Danish/Norwegian) are understood not as letters with diacritical marks but, rather, are entirely distinct and independent letters unto themselves. A substitution with a different letter is completely invalid and is a major spelling mistake.
    – J...
    Commented Mar 23, 2017 at 12:20
  • 8
    "the correct spelling is (almost always) the one on the publication itself" - I do not put a lot of trust in the capabilities of editors to maintain correct accents in the authors section (or anywhere else in the paper), either. Usually, the intended version can be guessed by looking at several papers of the author at different publishers, and possibly at their official website. Commented Mar 23, 2017 at 13:04
  • 25
    @O.R.Mapper "For citation purposes...", Part of the point of citations is so that a reader can find and read the rest of the work being cited. If the cited work bears the "wrong" name, and the citation corrects it, it would be harder to find. Commented Mar 23, 2017 at 13:47
  • 10
    @TomKelly "Not feasible"? If you're citing 'em, you've read 'em, right? So you can skim-read the author names for diacritics, and if there are any, check they exported OK. Reading the papers should be far more laborious than scanning the front cover for diacritics, and exporting and pasting the cite more laborious than checking pasted text for expected diacritics. So, how's this an issue, unless you're citing papers you've not actually seen? Also, is it normal in your field to forbid "et al." even for papers with more than five authors? Commented Mar 23, 2017 at 19:44
  • 4
    @PaulButcher: Indeed, there is a conflict of goals there. I doubt it is practically relevant, as the paper title, the venue, possibly the year of publication, and of course the DOI and embedded hyperlinks, if present, provide enough alternative anchors to identify a particular publication. None of those is as useful in giving credit, though, hence I prioritize correctly referring to personal names over optimizing discoverability when it comes to names. Also do note that this aspect is two-fold - misspelling the name to match the paper increases discoverability on the publisher's website, ... Commented Mar 24, 2017 at 7:29

This isn't a matter of ethics: it's a matter of respect. It is disrespectful to spell somebody's name incorrectly. It is doubly disrespectful to knowingly spell somebody's name incorrectly because you're too damned lazy to do it right.

  • 1
    Yes, I agree with you (and several others stating similar above). As I mentioned the the question, I have figured out how to include diacritics in my references in LaTeX and have done so for the names I am aware of. My question pertains to the massive task of identifying whether other authors are missing diacritics so I might miss some of them and it could be considered poor academic conduct if did.
    – Tom Kelly
    Commented Mar 24, 2017 at 11:09
  • 1
    I disagree. I am not sure what encoding the publisher is/will be using. Since accents are not within basic ASCII table I do not want to risk somebody's name (including mine) to be spelled with blank space or some bizzare glyph instead of accented letter. Therefore I use unaccented letters; just to be sure it won't get worse. If I am the one who rules the encoding I am reponsible for typesetting it correctly.
    – Crowley
    Commented Mar 27, 2017 at 12:52
  • 1
    @Crowley There are journals that require submission in ASCII? Everything I've seen wants either LaTeX or Microsoft Word or a PDF, and all of those can represent accents without difficulty. It's 2017, not 1987. Commented Mar 27, 2017 at 13:19
  • 1
    Yes it is 2017 and people still use MS word for tasks where other programs are far superior and MS Word is no capable of labeling images and tables right. It is 2017 but many people have computer knowledge and skill comparable to average person from 1950...
    – Crowley
    Commented Mar 27, 2017 at 13:47
  • 4
    @Crowley You said that you omit accents because you're worried about journal submission formats mangling them. I pointed out that none of the three main formats asked for by journals do actually mangle accents. You have said nothing that counters that statement. Please note that my comment was not an invitation to you to criticize any other aspect of Microsoft Word. Also note that, in 1950, probably only a few thousand people had used a computer at all. Anybody who has used a computer has more computer knowledge and skill than the average person from 1950. Commented Mar 27, 2017 at 13:54

I fully agree and can give extra input into @BurakUlgut's answer (+1), and @Ponts answer about the names in the paper itself. I want to reinforce the advice that you should use the names as the author's themselves put on the papers. This will prevent confusion in some cases.

In the country where I was born my name contains an accent (actually a non-latin1 letter, which is a variation of the letter l), and it is written that way on my birth certificate. Yet, since I moved out of my country of birth when I was a child and in the following two countries that I lived and worked as an adult I never used the accent because my parents omitted it from my entrance documents. On the single paper I have published to this date my name figures without the accent.

Now, had you gone through the trouble of checking my nationality and checking how my name is properly written in my country of origin you would make a mistake in the citation. Moreover since the name that would be written in the citation would not be possible to link to my name in almost any of my documents.

Yes, this is a strange corner case. But, I believe, that it illustrates well why the names on the paper itself are a good choice for citations; i.e. assumptions you make about someone's name may be quite wrong.

99% of my documents say Michal, I was born Michał, and people often make mistakes like Michael. (And I tell everyone to call me Mike to make things simple)

  • 3
    Thanks for sharing your take on it. Accents are all edge cases in English (e.g., loanwords and names) really. It is exactly these circumstances I was curious in hearing from people affected by it.
    – Tom Kelly
    Commented Mar 24, 2017 at 7:15

The purpose of a reference is primarily to allow others to follow along with your thought process and independently consult the materials you used to support your statements. As such they need to be accurate enough to allow your readers to find the original papers.

The second purpose is to be able to "traverse the tree of knowledge" by finding all other papers that cited a particular one. This means automated tools must be able to recognize the citation. Most tools strip accents - precisely because of this problem.

Everything else is courtesy - if people get offended because of a missing diacritical mark, that is unfortunate; but it is hardly a breach of ethics.

EDIT: Note that when I answered the question, it asked whether it was "ethical" not to get all the diacritical marks. The question was since edited to ask if it was "bad". It is clear that some people will take umbrage when you misspell their name; unfortunately, the same authors' names are not always spelled the same way in different papers, which frankly increases the level of confusion when you try to do it right (if you referenced Erdős here and Erdös there... is that the same person? People with average eyesight might not even be able to tell the difference. Neither might average proof editors. Only one of these spelling refers to a famous mathematician. But if you Google "Erdos" you will find him, either way.)


Generally it is standard practice to use as much of the original accent marks as the format allows. (For instance a 7-bit ANSI or 7-Bit ASCII Email is very limited in what characters it can display.)

It deemed respectful to use the added effort to use special computer tools to support the proper accent marks. Most word processors have these built in, but if not your OS should have a character select tool with "related characters" to find the various accented versions of a particular letter.

Additionally, you should check if there is a localized version of the research material for your region, some authors like to use localized versions of their pen-names, and in this case it would be rude to use the wrong pen-name, and citing a localized version of a resource can provide a lot of ease to the audience of your work when they are doing further research.

  • ANSI is a series of extensions for 7bit and 8bit text data streams. They include colorization and modifiers such as bold, inverse and underline. These standards and their successors are widely used in unix-like systems. There is a second set of 8 bit ANSI standards ussing 8bit text, called the ANSI codepage extensions. The windows implementation of this is slightly broken in that it just sets a single codepage instead of letting you switch it out. Commented Apr 22, 2017 at 8:15
  • ANSI is the predicessor to ASCII. The first standard was later adapted into ASCII with minimal changes. The later revisions include a 7 bit standard then a plethora of codepages for the "High 7bits". Commented Apr 30, 2020 at 1:58

You must log in to answer this question.

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