5

I’m currently reviewing a paper and thought it may be good to consult more people on this issue. I’m wondering where exactly should we include citations in article main text (in a science/psychology paper).

The paper starts with two paragraphs of intro with zero references in them with quite general claims. The logic behind it should be that later text will validate the claims. The same trick is used for literature-review paragraphs, where the first sentences make general claims/generalizations and the remaining paragraph seeks to validate them.

This system seems to me to be inconvenient for the reader as the remaining data given is more specific than the claims, and it needs work to see whether they actually cite enough studies to justify the earlier generalizations. Sometimes it seems they don’t but the style itself makes it difficult to immediately perceive.

I am used to the style that citations should be introduced whenever you introduce information (especially general claims). I am wondering if anyone can confirm that the other way works and is common also in practice, where the general claims are followed by more specific validation.

It seems appropriate for me that the sentences could really be left uncited if they are common knowledge and for introductory purposes. But not if they make claims to be built upon. The authors do think that these are claims, they just present the summary before the facts.

For example (made up of course, and to be read by someone who doesn’t know tea):

Tea leaves are generally green. They look green to most people and animals. For example they have been found to be green in sunlight (Authors 2012), and particularly the leaves of carnelia sinensis has been found to be very green (Other Authors 2011). [A few more specific cases …]

I’m sure academia is full of different styles of writing. I’d be happy to hear of different ways to approach presentation. Any style guides proposing this that you could point me to?

  • 1
    For better or for worse, I see this pattern frequently in Computer Science. It does not really seem to be perceived as a big issue to anybody, as long as the authors then do not go ahead and build upon the general and uncited claim rather than the specific cited one. – xLeitix Aug 4 '16 at 11:39
  • Ok, thanks for the info, it's good to know that it is common enough elsewhere. Sounds sensible enough too. – puslet88 Aug 4 '16 at 12:36
  • 1
    This is a good question. In my experience, I have usually preferred saying things like: "Johnson (2000) found that x<5 by reticulating the spline and Smith and Jones (2013) found that x>2 by polarizing the ion matrix according to McWhooferton's (1980) pattern. From this, it is evident that x is somewhere between 2 an 5, exclusive. In this paper, I will show why we can be 90% confident that x is within 0.2 of 3.7." That is, one starts with the individual claims and then builds them up to show why the broad claim is true. – Columbia says Reinstate Monica May 31 at 17:00
5

Certainly general statements that are common knowledge do not require citation. Particularly if the general claims just help to provide a framework for thinking about things (context), are reasonable, and are not truly needed for the paper.

However, different people have different opinions about what is common knowledge, or is reasonable, or is important (say in putting the work in context), and part of your job as a reviewer is to point out where you think citations should be added or assertions should be questioned. The authors always have the option to disagree, but at least if you make a suggestion, they should think more carefully about the validity and evidence for such an assertion. (cf. my answer https://academia.stackexchange.com/a/66329/19607)

In your example, you might ask for evidence that they "appear green to most people and animals," or just point out (politely) that that's a stupid statment.

As another example, here's something I wouldn't expect a reference for:

The only known techniques to attack Problem A are Method X and Method Y. Method X was pioneered in [cite 1] and Method Y was recently applied in [cite 2].

It's hard to give a reference for the first assertion, and maybe the reviewer will be aware of a Method Z and can point it out. Or maybe there is a Method Z but it got neglected by the authors and the referees. Or maybe the referee's not sure. This is why some people might prefer to be more modest and say "The only techniques we are aware of..." but people write in both ways, and as a referee I wouldn't say anything unless I actually thought the authors were neglecting something.

  • In your example, I think what the author means is "the only techniques known to me are X and Y," not "the only techniques known to humankind are X and Y." I would expect a proof or citation if the author does mean the latter. I see many authors put a cliché like "as far as the author is aware" or simply add something like "it appears that" to avoid this, which sounds to me better than making too strong a claim if taken literally. – Yuichiro Fujiwara Aug 4 '16 at 14:10
  • @YuichiroFujiwara I meant the author strongly believes no other methods (that have had at least partial success) are known by the scientific community (where you can take "known" to mean published and/or publicly announced). While I often try to qualify sentences like this to say "to my knowledge," I sometimes don't even though I of course can't prove someone hasn't published a result I'm not aware of. – Kimball Aug 4 '16 at 16:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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