I am wondering if it is okay write something like:

"Many scholars hold the view that this is like that" and then give one example reference (e.g. Gans et.al. 1965) ?

The reason I am wondering because it is generally accepted and no one seems to disagree on that. Is it better (1) not to reference at all, (2) reference the most original paper and skip e.g., or (3) mention few?

4 Answers 4


If something is considered common knowledge in the field (for instance, Newton’s second law in a physics paper), then no reference is required. If it’s been developed multiple independent times, though, I’d use more than one reference to establish it.

  • 1
    However, Newton's second law in a paper from some other field: do include a reference. For example a standard physics textbook. One reference is enough. However: also include the page number in your reference, do not merely cite the whole book.
    – GEdgar
    Apr 11, 2018 at 13:53
  • 3
    @GEdgar I can't imagine citing a textbook for Newton's second law in any non-school context...
    – Bryan Krause
    Apr 11, 2018 at 17:49

I have used the following construction:

Many scholars hold the view that x is like y (Smith et al. 2000; Brown et al. 2001; and many others as cited in Stone et al. 2005).

  • 1
    Usually I try to reuse citations. If I already cited a paper which also states something more obvious, I cite the paper a second time instead of adding a new entry to my references just for a trivial statement.
    – usr1234567
    Apr 12, 2018 at 20:34

Here are a few ways you can keep the citation with just that one resource. I'd rewrite to avoid the "many scholars hold the view" part (which you would need to defend by citing multiple papers). Instead, you can emphasize the idea itself.

If you're setting it up to disagree with it:

It has long been held that X is like Y (see, e.g., a comprehensive argument in Gans et al. 1965).

(Though if you're getting ready to argue against it, you will want to go into further detail and cite several different sources.)

If you're just trying to get that fact in there as background you're building on, you can say:

It can be useful to use Y as a model for X (see, for instance, a treatment of this question in the context of Z in Gans et al. 1965).

X can be thought of in the same way as Y (e.g. Gans et al. 1965).

X resembles Y (e.g. Gans et al. 1965).

All of these formulations shy away from asserting X is exactly like Y, or that X is a type of Y, but they get the point across and give people further reading.

Given all the time in the world to track down every bit of nearly-common knowledge, you would find and cite the original scholarly reference for the idea, AND (if needed) recommend a later paper that would help the reader because it is easier to read or is written in a context similar to the context of your own paper or it uses modern notation. Readers who are aware of the original source might be amused/disturbed to find only a "lesser" reference there without explanation.


If you are certain that several scholars hold the view, then presumably you know where they said that, and you should cite several sources. If you are not certain, you should look for sources.

Be careful to avoid propagating untrue information and wrong citations. For example, Friedman & Koller (2003), a machine learning paper about learning Bayesian network structure, says

in small domains with a substantial amount of data, it has been shown that the highest scoring model is orders of magnitude more likely than any other (Heckerman, Meek, & Cooper, 1997).

But Heckerman, Meek, & Cooper (1997) only showed the claim to be true in one specific case. Friedman & Koller's claim might be true, but their implication that Heckerman, Meek, & Cooper (1997) showed it to be true is not.


Friedman N, Koller D (2003), Being Bayesian about network structure: a Bayesian approach to structure discovery in Bayesian networks. Machine Learning, 50, 95–125

Heckerman D, Meek C, Cooper G (1997), A Bayesian approach to causal discovery. Technical report MSR-TR-97-05, Microsoft Research

You must log in to answer this question.

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