The Review Article: An Unusual Beast

This post is about review articles in physics. Below I will put down my philosophy about what reading and writing such articles involves.

1. What they are: Review articles are useful but somewhat unusual objects, in that they do not represent original research, like most published papers, but rather a summary of (usually recent) developments in a field.

   
   

2. Why they are useful to the reader: Review articles are useful to researchers looking to get into a new area of physics. This may be a graduate student trying to learn a field for the first time or a seasoned researcher looking to expand their expertise. Review articles are useful in these cases as they typically include i) an introduction to the basic ideas ii) a modern understanding of the relevant concepts and iii) a survey of the relevant literature, e.g. the main theoretical formalisms, the milestone experiments, etc. No review article is fully comprehensive, of course. But a good one can form a starting - and reference - point for systematic exploration.

   
   

3. Why they are useful to the writer: The review article allows its author(s) i) to revisit the roots of their subject, ii) assemble their thoughts on the latest ways of thinking about the subject iii) survey the landscape (which clarifies the gaps and challenges in the field) iv) impose their own interpretation and understanding of the material, etc.

   
   

4. What kind of journals publish them: A variety of peer-reviewed journals publish review articles. In physics, some journals are almost entirely dedicated to publishing review articles, such as the Reviews of Modern Physics (which also publishes colloquia and Nobel prize lectures). However, others, like the Reports on Progress in Physics and Review of Scientific Instruments publish both review as well as research articles.

5. What kind of topics are appropriate: The most consequential review articles concern themselves with an area of physics that has emerged or produced energetic advances over the last (half a?) decade or so. A number of leading groups around the world usually contribute to such advances, a flurry of papers appear in the high impact factor journals on the topic, a number of these papers have hundreds if not thousands of citations, etc. etc.

   
   

   Having said that, some professionals also write review articles focused around their own contributions. The relevance of such reviews depends on how impactful their research is perceived to be by the community. Generally speaking, if there is something that only you - and perhaps your friends - are working on, then the community might not be waiting eagerly to hear that work summarized. Unfortunately, many people try to game the system: by the very fact of being reference points, review articles are highly cited, and writing them is a way to raise your own metrics (such as the h-index). I am not in favor of such activity.

   
   

6. Who should write them: To my knowledge, review articles are published mainly through two routes. The first route consists of a journal inviting some professionals to write on a topic that the journal itself has identified as important and timely. The second route consists of groups of professionals who have assembled a review article by themselves and are looking for a venue to publish their paper. The rejection rate is much lower for the first category, though I have seen some such papers be eventually rejected if the peer review is largely unfavorable.

   
   

   In all cases, the review article is mainly written by professionals (academics or industrial researchers) who are distinguished in the field. What does it mean to be distinguished in the field? This is a subjective question. I like to say that the person i) must have worked in the field for at least 10 years ii) published at least 10 papers iii) at least 5 of these papers must be in high impact factor journals and iv) at least 5 of these papers must have more than 100 citations. All these conditions and numbers are to an extent arbitrary, but they give a flavor of how high the bar should be, in my opinion.

7. Who should not write them: I should perhaps say 'who should not expect to write them'. It should be clear from my exposition above that undergraduate and graduate students and postdocs should not expect to generally figure as authors in such articles. This is because they are early career academics and not leaders of the field.

   
   

   Of course, there are exceptions: I have been part of review articles where a postdoc has been added to the author list as the professors are too busy to draw figures, get permission from journals to reproduce diagrams/data, add references, or knock the prose into readable form. A much more distinguished exception to the rule was Pauli, who at 21 years of age, wrote a 200-page review of relativity that dazzled even Einstein.

   
   

   But generally I consider it inappropriate for even postdocs to suggest to professors that the time for a review article is due, because i) the professor has a much better idea of the timing and appropriateness (they might be waiting for verification of theoretical predictions, or better theories, or a more impactful era) ii) the prof might have to negotiate with collaborators (academic politics is often involved) to decide who should be the authors iii) the postdoc has not had a decade to make contributions or the experience or knowledge to call time, and many other reasons.