9

So the title of the question is quite broad, I've seen a few of questions in the site regarding this particular topic (quite a few actually, so let me know if this is somehow a duplicate).

My problem is the following, I am working on my PhD thesis, and so far me and my supervisor have collaborated in a few articles (that somehow constitute the body of my thesis). Right now I am working on a particular article (on which I've made all the work). The fact is that my supervisor, and apparently all the members of my faculty tend to put authors in alphabetic order. The fact is that my last-name starts with a "T" so by using their logic I will almost surely end-up being the second\third author.

I don't feel this is actually fair, since after some search online I've seen that in general the order of the authors reflects somehow how much work was done by each one. So it seems strange for me to be like "coauthoring" my own PhD thesis.

What should I do? I don't want to appear pretentious, or rude in front of my supervisor, but I don't think it's fair for me either.

Do you have some advice on what to do?

In case it's relevant my field of research is pure math.

8
  • 7
    I believe the authorship order convention somehow depends on the field. From this answer, alphabetic order is the convention for pure math
    – Nobody
    Nov 18, 2020 at 9:28
  • 3
    If you did all the work, your supervisor should not be included as an author. It's fairly typical for Math PhD students in the later stages of their thesis to write solo-authored articles. Nov 18, 2020 at 9:37
  • 2
    It's a bit, but not massively unfair. A bit unfair because the first name gets a bit more visibility, especially in citations. Not massively unfair because people in your field (that will assess you in your career) understand the limitations of alphabetical order and know that it doesn't say anything about the actual contributions. Nov 18, 2020 at 9:44
  • 1
    @Alchimista Yes, I was responding with the context of the stated field of OP in mind. Nov 18, 2020 at 9:47
  • 1
    @lighthousekeeper sorry, didn't notice that
    – Alchimista
    Nov 18, 2020 at 9:47

5 Answers 5

25

The convention in pure math is to list authors in alphabetical order.

Since this is generally known inside pure math, readers will not draw any conclusions from the author order about the relative importance of contributions. In fact, my impression is that most pure mathematicians prefer to avoid discussion of relative importance of contributions altogether, with the exception of occassionally giving more credit to their coauthors in slightly informal settings.

[As a side note, Author contribution statements seem to be unfeasible for widespread use in math. While there are collaborations where X proved Theorem 1 and Y proved Theorem 2, it is in my experience far more common that X and Y talked repeatedly over a long period of time, and at some point the proofs materialized.]

Admittedly, being listed first will mean being noticed a little bit more. However, in line with the convention about author order, the "X et al"-style is typically avoided. It seems far more common to refer to unwieldly large groups of authors by initials (eg GKSSW proved that...).

The place where your advisor is supposed to express how overwhelming your contributions were is in your letters of recommendation.

That said, in pure math it is also convention that merely advising a student does not establish authorship. It is even a common sentiment that advisors should be very reluctant in appearing as a coauthor on an advisor/student paper.

3
  • 3
    An important point is that people outside of mathematics (e.g. Deans and other higher level administrators) may be evaluating faculty in pure mathematics without understanding this convention... Nov 18, 2020 at 21:19
  • @BrianBorchers your comment is very accurate, and indeed even if I didn't state it in the body of the question, I do math research but in an Statistics department!
    – Chaos
    Nov 19, 2020 at 11:34
  • 2
    @Chaos If you come across someone who seem like they might not know how these things are done in mathematics, you can link them ams.org/profession/leaders/CultureStatement04.pdf Nov 19, 2020 at 15:05
11

"What should I do? I don't want to appear pretentious, or rude in front of my supervisor, but I don't think it's fair for me either.

In case it's relevant my field of research is pure math."

My recommendation is to accept that alphabetical authorship is the ubiquitous convention not only in pure math but also in most areas of math and computer science. You did the right thing by asking here instead of complaining to your supervisor.

If your concern is that you might in the future want an academic job in physics or engineering or a non-academic job where the recruiters might make the mistake of thinking that you were a minor contributor to all those papers, then you can always put something like this in your CV at the top of your publication list (this is taken from my friend's academic webpage):

enter image description here

2
  • 1
    This is very interesting, I will for sure add this to my CV!
    – Chaos
    Nov 19, 2020 at 11:35
  • Maybe acknowledge the original author or write your own version.
    – Nik
    Nov 19, 2020 at 14:42
6

Let me suggest that in pure mathematics the conventions are that a student is usually permitted/encouraged to write sole author papers. And for properly done joint work, the convention is to use alphabetical order for authors. Mathematicians will understand this though people in other fields might not.

But, you can also include a short "contributions" section in the paper detailing who did what and how the various people contributed to joint work.

My suggestion, though subject to your advisor's approval, is to write this as a sole author paper. But the reason for the advisor's approval is really just political, not ethical. If they oppose it and thereafter make your professional life difficult, then you gain nothing.

I studied math, but taught CS for the most part. I had a few doctoral students. I'd have been amazed if any of them thought it would be a good idea for me to be a co-author of their work, though I guided some quite closely. A polite thank you somewhere for any help I gave is enough. I think this is pretty standard in mathematics and in theoretical CS.


Of course, there may be some places in the world where the conventions, even in math, are different and unknown to me.

5

In fields where alphabetic ordering is the norm, non-alphabetic ordering is sometimes used for indicating that the first author contributed much more. (See for example https://arxiv.org/abs/hep-th/0510032). Asking to be first author would seem reasonable in your case.

This said, author ordering is a rather coarse way of indicating who did what. Some journals now make "author contribution statements" mandatory, see this Nature editorial: https://www.nature.com/articles/4581078a

7
  • 1
    I don't know if a single example makes a good case for "regular use". For example, a quick check of the 2019 and 2020 editions of Annals of Mathematics shows that all published articles used alphabetic ordering. Nov 18, 2020 at 12:59
  • 1
    In theoretical physics there are many other examples. I would estimate that the fraction of papers with non-alphabetic ordering is of the order of 1/10. Out of 18 papers announced on arxiv/hep-th today there are 10 alphabetic, 6 single-author and 2 non-alphabetic. Nov 18, 2020 at 13:26
  • Interesting. Potentially there are also differences between the areas that default to alphabetic ordering. Nov 18, 2020 at 13:28
  • 4
    I think it's more accurate to say that non-alphabetical ordering is sporadically used to indicate that the first author contributed vastly more than the second/others. At least in the parts of math and theoretical computer science where I work, I'd estimate that less than 1/1000 (sic, thousand) papers with multiple authors list them non-alphabetically.
    – JeffE
    Nov 18, 2020 at 14:43
  • 1
    The comments in the thread are a classic example of people finding it ridiculously hard to believe that people in different communities behave differently. The fact is that different academic fields have wildly different conventions. @SylvainRibault says that in the field of theoretical (high-energy/particle) physics, alphabetical ordering is the norm, but occasionally violated for the sake of indicating relative contribution. You can say it's different in your field, but it doesn't mean that he is wrong.
    – Aqualone
    Dec 2, 2023 at 20:02
-2

(edit: my take from the Computer Science community)

I'll start by saying that every research group has their own conventions. Some allow every member of the group to be coauthors of every paper produced within that group. Others use sophisticated systems like counting the lines of text written to establish what is the highest contributing author, then establish the list.

No matter what style your group has, its conventions should be first accepted as valid, since they were probably established a while ago. If they seem implausible, they should be discussed or challenged. From my perspective, it's more important, especially at your stage, to learn the discipline of this job, rather than trying to change its structure.

Now coming to your academic production: regarding papers, I'm of the opinion that PhD students work together with their advisor. Advisors should always contribute to their students' work, and a PhD student should always include them in the authors' list. I am actually happy with the grey area of contributions: since you and your advisor discuss about progress, next steps and ideas, even though he/she might not write a single line in your paper, I think that they should be still considered coauthors. Their contribution, even if only reading the paper and asking to change here and there, is indeed a contribution. I know there is a lot of discussion about that on other platforms, but this indeed is my own opinion.

I see your point with alphabetical order: my surname is with "V" so I had the same issue in the past. It sucks but I worked my socks off to be considered the first author, but at times I did not succeed.

On the other hand, your PhD thesis will be submitted as your own work. I do not know of cases where the final thesis had any another name, apart the candidate's.

6
  • 5
    I'm of the opinion that PhD students work together with their advisor. — You're welcome to your opinion, but you're wrong. I don't let my PhD students graduate until they publish at least one paper without me.
    – JeffE
    Nov 18, 2020 at 14:46
  • Wow... that answer has turned into a lot of down votes. If people could explain why they do so, it would be appreciated.
    – ElCid
    Nov 18, 2020 at 14:53
  • No I said: in my opinion phd students should always work with their supervisor. I don't have a universal truth, I offer my attitude to PhD supervision. I welcome your difference of opinion and I don't claim that yours is wrong
    – ElCid
    Nov 18, 2020 at 14:56
  • 7
    See the other answers! In math (and related fields) it is extremely common for senior PhD students to work independently from their advisors. Arguably the goal of any PhD program (at least in these fields) is to produce high-quality independent researchers. From a practical standpoint, PhD students who never publish independently from their advisors don't develope independent research reputations, and therefore have a significantly harder time finding academic jobs. (That's why I require my students to publish without me.)
    – JeffE
    Nov 18, 2020 at 14:57
  • 4
    tl;dr: Forcing your students to include you in all their research actually hurts them.
    – JeffE
    Nov 18, 2020 at 14:59

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