Refuted papers: do they belong in a literature review?

Question

I'm writing a detailed review article (for submission for publication) that aims to survey the various proposed explanations for a particular biological phenomenon. I'm aware of one explanation (that was proposed 20+ years ago), but a number of papers since then have pointed out fatal flaws in its logic. The original paper was published in a respectable journal and has been relatively well cited, since many papers on this phenomenon include some form of survey of the proposed hypotheses. However, it has not received any (published) support, experimental or otherwise.

At what point can we, as researchers in the field, consider the matter closed, and cease to refer to the refuted hypothesis? It seems a waste of space to spend a paragraph detailing one argument, only to spend the next paragraph explaining why it is flawed, and then not referring to it again. On the other hand, does it damage the credibility of my review if I omit one of the hypotheses without any justification?

Answer 1

Based on the background you gave, particularly that you are writing a detailed review article, then I think you should definitely include this article.

Part of the function of a good review article is that it documents the history of scholarly thinking on the topic of question. From your explanation, this article presented an idea so apparently true that not only did some intelligent researchers study it and present it as fact, but it even passed peer review in a high-quality journal and was cited by subsequent researchers (though, of course, many people might have cited it to contradict its results). This indicates that it has played an important role in the evolution of knowledge in the area.

Think of it this way: suppose some researchers read your review article that omits this idea, and then they independently get that same idea as their own original idea. Without knowing that it has already been refuted, they waste their time conducting a study based on their ideas, and probably get negative results, to their disappointment. (Or worse, they get positive results, submit them for publication, and then get slammed in peer review for not knowing the literature.) By documenting the full body of knowledge, you can spare these hypothetical researchers all that pain.

Although my area has nothing to do with biology, I still remember in secondary school biology class learning about spontaneous generation as an early theory of genetics. I was deeply impressed by the explanation of how such a popular theory was eventually refuted by better science. That case was an important educational milestone in developing my scientific thinking. By including the refuted work in some detail (in correspondance with the amount of scholarly work spent developping and then refuting it), you could help your readers in a similar way.

Answer 2

Given that you are writing a review it is probably better to mention the old paper: if it was so well cited, readers of your review are likely to come across it and it will help them greatly if your review says they are better off looking at more recent theories.

With that being said: it is your review so you can write it in the way you like. If reviewers think you should have mentioned the older paper they will let you know and you can still include it in a revised version.