In my research field (operations research), several works have been published in the last couple of years by different authors that cover most of the recent developments in the field. I’m writing a review paper and even though I read and used a couple of papers dated as far back as the 90s (mainly for my introduction and review), I want all references to be from 2010 or later.

I would like to write something like:

Ref. 3 discussed branch & bound and dynamic programming approaches for KP, MKP and some of its variants.

Reference 3 is Kellerer, Pferschy, and Pisinger (2004). Another paper, namely Puchinger, Raidl, and Pferschy (2010) shares one of its authors with Reference 3 and also cites it.

My question is: Instead of citing Reference 3 as it is, can I cite Puchinger, Raidl, and Pferschy (2010) because it is more recent?

  • 8
    I want all reference to be dated from 2010-2019 --- Just a passing thought, but wouldn't some readers suspect that you were not very thorough in your review if all your references were only from the past decade? Mar 4, 2019 at 15:46
  • 3
    To follow-up on my comment, I spent a few seconds just now googling for a survey paper (googled the phrase "survey paper" along with a random math topic, "convex geometry") and I found this paper published in 1999, and I tried to imagine how thorough the coverage would be if all references before 1989 were omitted. (Not very thorough was my conclusion.) Incidentally, note that this survey paper is much narrower in scope than "operations research". Mar 4, 2019 at 16:24
  • 3
    Still, why "instead" when you can cite both? "This first appeared in [1]; see [2] for a modern exposition" is a standard way to proceed when you want to cite a newer source [2] along with an old one [1]. Mar 4, 2019 at 16:54
  • 1
    Is the 9 key on your keyboard broken? Mar 4, 2019 at 17:09
  • 4
    Before anyone can give a meaningful answer, you need to answer a question: why? Why do you "want all references to be from 2010 or later"?
    – user68958
    Mar 4, 2019 at 17:16

2 Answers 2


I would recommend against this as it seems artificial. The year you give isn't some magic number and citing only recent work doesn't give you any advantage.

But there is a deeper reason in this case, I think. Just because a new paper cites an older one doesn't mean that the newer one advances the older one in every relevant direction. Another way to think of it is that the ideas in the new paper don't actually "contain" all of the ideas in the older one. Someone wanting a good overview of the field - the purpose of a review article - will probably need to go see what those additional ideas are, but if you don't point to them their only pointer is indirect through the paper you do cite. This makes the serious reader's job harder.

You seem to be giving an annotated bibliography, I see no reason to truncate it. The annotations, if well done, will help a reader decide whether to go to any given paper you include.

  • I suppose that one could write an 'update' review, covering just the last $n$ years worth of work, but that point should be made explicit as well as having citations of the last few major review papers in the field. Slightly different focus of the paper, and it would be written a bit differently.
    – Jon Custer
    Mar 4, 2019 at 18:57

Cite them both. Covers various aspects.

You must log in to answer this question.

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