Recommendations for citing papers

By Roberta Millstein

Following on Helen De Cruz's excellent Why we should cite unpublished papers and some recent reflections of my own while refereeing, I thought it might be helpful to compile a list of suggestions for when to cite (now that we know that our citations should include both published and unpublished work):

1. If someone has provided a way to understand a certain debate that had not been recognized before and you find it useful to present the debate in that way, you should cite them.
2. If someone has provided conceptual distinctions that you are using in your paper, you should cite them.
3. If someone has done the work to find and explain a case study and you want to refer to that case study too, you should cite them.
4. If X has developed further the ideas of Y, you should cite both X and Y. 

I do think, however, that one can "over-cite" or "mis-cite."  An example of over-citing is listing every paper that is on the same topic as yours.  I am not calling for that.  But content that is relevant to your argument should be cited; another reasonable practice is, I think, to cite a paper that contains lots of good citations in it (e.g., "See Z and citations therein.").  An example of "mis-citing" is assuming that because X presented Y as saying A, you can safely present Y as having said A.  No.  You need to go read Y yourself.  Too often misrepresentations are perpetuated throughout the literature.

It seems to me that we need not simply wring our hands about citation practices in our field when it is in our power to improve them.  We can do better ourselves (I include myself in that "we") and we can insist on better practices when we are refereeing papers.

I welcome suggestions for other best practices while citing.