40

I am wondering if it is a bad thing to cite a paper in a scientific publication for a trivial or irrelevant reason.

The specific instance I have in mind is the usage of certain terminology. I basically want to make a point along the lines:

"We define a set to be flabby if it obeys conditions X, following the convention in [Smith]. We note that other authors (e.g. [Jones]) also require flabby sets to satisfy a condition Y".

Now, it is clear that if you use some definition (which is not very classical), it should be attributed, or at least a reference should be given to some reasonably good introduction to the topic. In this case, the reference the paper of Smith does the job just fine. But what about Jones? The only reason for citing him is that he happens to be using a different convention than I. On the other hand, I cannot convincingly make the point I want to make without some reference; and I'm concerned that I might be confusing the reader if I don't make this point. It could be that Jones gets an extra epsilon of recognision because of one more citation to his paper, which I personally don't mind at all. But it is maybe slightly weird that I cite a paper which I am not, strictly speaking, using.

I suppose this particular case is not really that important. What I am would really like to know - although that's perhaps too vague for SE - is whether it is generally OK to cite papers just because it is convenient for me, without worrying about whether I actually use the results of that paper in my work.

(If relevant, my field is pure mathematics.)

  • 4
    I agree with the answers that you should include a reference for the convention that you don't use, but I would add that you should find a reasonably well-known source for that convention, for example a standard textbook. If the convention in question is used in well-known sources but [Jones] is some obscure paper, then your choosing it as your reference will look like an attempt to give Jones an undeserved citation. – Andreas Blass May 31 '15 at 22:38
  • I love sheaf theory and the nomenclature thereof. – S. Kolassa - Reinstate Monica Jun 1 '15 at 12:14
  • @AndreasBlass If you turn this into an answer, I will vote it up. – jakebeal Jun 6 '15 at 19:24
58

Even things that you aren't directly using may be quite relevant for establishing context.

The only thing that ever keeps me from being generous with citations is page limits. Otherwise, any citation that fits well with the flow of the scientific narrative and helps place your work in the context of related work is good for everybody involved, and I see not reason not to err on the side of inclusion.

  • 2
    Thanks! Erring on the side of inclusion was in fact my first instinct - just wanted to confirm it's not a bad practice to do it. – Jakub Konieczny May 30 '15 at 19:03
  • 8
    Exactly. Context. Explanation. There are larger purposes to writing than any competition for "novelty" or "impact factor". Clear, helpful writing that doesn't play games is excellent. – paul garrett May 30 '15 at 22:16
36

jakebeal is right on, both that you should include the citation, and why. I want to respond to something you said:

But it is maybe slightly weird that I cite a paper which I am not, strictly speaking, using.

because it's part of a recurring misconception about how citations should be used. Papers shouldn't be the bare minimum necessary to claim priority on whatever's being done in them; they're supposed to be written to help other people understand what we've done. There's nothing wrong or unprofessional about including content which "merely" makes the paper easier to understand. What you're describing is squarely in that category: pointing out the alternate convention helps some readers avoid confusion, and providing a citation both supports that this is an alternate convention and gives some hint where in the literature the alternate convention is found.

  • Yes, good: claiming priority ought not be the deciding feature. :) – paul garrett May 30 '15 at 22:17
16

A long time ago in a galaxy far, far away, I thought that citations should be kept to a minimum: really, I didn't want to burden the reader with references to trivial or generally known stuff.

Along the years, however, from discussions with colleagues and from received paper reviews (e.g., countless please provide references for the notation/equation you employed on p. X), I've grown convinced that what I once thought trivial or generally known is not really so: after all, even people working in the same field can have different backgrounds (e.g., physicists and engineers don't speak the same language).

Thus, today I think that one should provide as many references as possible: to better explain the context, to provide extended information on notations and background theory, and to outline existing differences in notation and terminology (as in your case).

So, be generous: you'll never know who your reader actually is, and providing more references won't hurt.

3

Answering to your particular case: Your citing Jones is only confusing to the extent that it invites the question of why you prefer Smith's definition to Jones'. If you can explain this preference (briefly), you defend your approach against potential criticism and, doing so, have a better reason for citing Jones in the first place.

  • 8
    I don't think this is really an issue in mathematics. When there are multiple definitions of a term, the choice is usually a matter of convenience, more than anything else. For example, the natural numbers may or may not include zero. It would be silly to use the definition N={0, 1, 2, ...} and then write N\{0} all the time if you mostly needed to exclude zero. – David Richerby May 30 '15 at 19:22
  • 2
    that makes sense. the issue then is field-specific. in political science, i couldn't go ahead and use a non-conventional definition of say, advocacy coalitions, without justifying this decision. – henning -- reinstate Monica May 30 '15 at 20:51
  • 1
    Still, a few words explaining preferences about conventions would be helpful to the reader, surely!?! – paul garrett May 30 '15 at 22:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.