I need to use a formula, found in a book "A". The book itself refers to an academic paper from the 50s, "B".

So when I use the formula, who should I reference?

We use the formula given as g = 4x^2 ("A", 2001).

We use the formula given as g = 4x^2 ("B", 1950).

  • 4
    Why not both? One to show priority, the other to show an easy place for the reader to find it.
    – GEdgar
    Commented Jun 2, 2017 at 17:55
  • 2
    It is best to cite both. But first you need to look at the paper from the 1950s yourself, to check that it did actually say what you think it said. Otherwise you might propagate wrong information. See my answer here.
    – user72102
    Commented Jun 2, 2017 at 19:41
  • I asked a similar question some time ago : academia.stackexchange.com/questions/68979/… The situation is not exactly the same, but maybe it could be useful.
    – Arnaud D.
    Commented Jun 2, 2017 at 21:35

2 Answers 2


You should try to cite primary sources whenever possible; in this case, that means citing the original paper "B".

However, as GEdgar says, a textbook may be an easier source for people to find, particularly if that book is the de facto standard text in your field. It is therefore a good idea (and perfectly acceptable) to cite both "A" and "B", for example:

We use the formula given as g = 4x^2 ("B", 1950, "A", 2001).

Personally, I prefer to be more verbose and might write something along the lines of

The formula we use is given by Author B ("B", 1950) as

g = 4x^2.

This was later used [confirmed/ expanded on/ usefully re-derived/ etc] by Author A in their book "A" ("A", 2001).

  • 2
    I particularly like the more verbose approach you suggest, since it makes the original authorship of the idea clear and also allows clear credit to the contributions that may have come after. In terms of usefulness, including multiple 'important' sources for a formula make is easier to later track down related works.
    – abase
    Commented Jun 2, 2017 at 20:47
  • 3
    Yes, but please do make sure you actually look at both the sources, to check that they say what you think they say. Don't trust other writers to have done correct citations.
    – user72102
    Commented Jun 2, 2017 at 21:12

Important and/or well-used equations/relations/results have a certain evolution to their citation pattern.

  1. First their original authors are referenced.
  2. After time has passed and it has appeared in several textbooks, people tend to just quote the textbooks, simply because these are much more accessible than an old paper.
  3. After it becomes a basic, people stop giving a reference altogether and treat it as god-given (who quotes Cooley & Tokey 1965 for the fast Fourier transform).
  4. Some relations acquire the name of their author (some even acquire the name of somebody else), such "Maxwell's equations": nobody quotes

    Maxwell, James Clerk (1865). "A dynamical theory of the electromagnetic field". Philosophical Transactions of the Royal Society of London. 155: 459–512.

    Many fundamental relations that form the basis of contemporary science have been developed by scientists in the past century, but rarely is appropriate credit given.

In your particular case, point (2) above appears to apply, but in the end this is a matter of taste and personal judgement.

You must log in to answer this question.

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