For example, take the sentence: "Gaussians are widely used in mathematics & statistics".

At this point there probably tens of thousands of papers that use Gaussian functions in some way (that's not even counting the papers in other fields like the social sciences that make use of Gaussians). It's impractical to cite them all. The sentence still calls for at least some citations though. How does one choose which papers to cite?

  • 2
    Not a fortunate example as it does not require citation. – Alchimista Jan 28 at 9:49
  • 3
    Perhaps a better example would be "deep learning has set new benchmarks on many tasks over the past decade." In my mind, this is a self-evident statement that does not require citation. Yet, most papers do have citations after this sentence, sometimes more than 10 (and, in one case, more than 70). – cag51 Jan 28 at 17:20
  • Do you have the option of citing a recent review paper and "references therein"? – terdon Jan 28 at 18:08
  • 1
    Not duplicate, because the first answer in that linked question says "If you think it is common knowledge, wait for the reviewer to ask for a reference, before going to all the trouble of sorting out the history", and in this case the reviewer has already asked for references. Voting to reopen. – Allure Jan 28 at 22:05
  • Not a duplicate, should be reopened. The example in the OP is not a good one, but there are other general statements that are not common knowledge. – Snijderfrey Jan 28 at 22:12

Either none at all if you can assume that the most likely audience of your publication is aware of that as if it was a basic fact of life.

Or recent reviews that provide a wide overview of Gaussians used in mathematics and/or statistics.

I searched for TITLE: (Gaussian) AND DOCUMENT TYPES: (Review) AND YEAR PUBLISHED: (2015-2021) in Web of Science. The query led to 56 results, many of which have accrued considerable citations and thematize mathematical or statistical issues. Perhaps you could do a similar search and pick some suitable papers to cite.

  • Indeed, in general it doesn't even have to be a (review) paper: in some cases there may be a standard monograph or textbook that provides a broad overview that you can cite. – Lou Knee Jan 28 at 21:28

If you really want to write such a general statement: Look at the statement and think about what it means. The references should provide evidence that the statement is indeed true. So in your example, you need a selection of references that show that "Gaussians are widely used in mathematics & statistics" (sorry, it is as simple as that). This selection does not need to be complete, nor must it be 100% representative. A few more or less random, but fitting references from publications using Gaussians within the mentioned topics suffice, maybe from different time periods and different (well-known) authors. I would go against citing reviews only, at least a few primary sources should be mentioned.

However, it makes more sense to be more specific and to give some more context. For example you could simply write:

Gaussians are widely used in mathematics & statistics, e.g. in topic 1 [citation 1], topic 2 [citation 2] and topic 3 [citation 3].

In this case, the selection of references is a little less random which is to be preferred.

Your example is not a good one, it is common knowledge and does not require citation. However, I treated it just as an example for a general statement, and of course there are general statements that are not common knowledge.

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