On writing a scientific paper

(Part 4 of 4)

How long should a paper be?

We maintain that a research paper should be as long as necessary — but no longer. We also stress that the length of a paper has no correlation whatsoever to its significance. We use as prime example the classic paper of Watson and Crick (referred to above) on their model of DNA. It is listed as covering two pages, but only because the papers in Nature then were printed immediately one after the other. In fact, the Watson and Crick paper could easily be rearranged to fit in a single page. That very short article and other papers along the same vein (also very short papers) won Watson and Crick the Nobel Prize. Another example of a short paper with great significance is the “Letter to Nature” by Köhler and Milstein in 1975 (Nature 256:495-496) on the production of hybridomas (the fusion of a myeloma cell and an antibody-secreting cell to produce large amounts of antibodies with a predefined specificity). This “Letter” also won Köhler and Milstein the Nobel Prize.

Reviewers

Reviewers are supposed to be fellow scientists who are familiar enough with the field so as to be able to render a correct opinion on the study being presented. In other words, they are “peers” of the authors. “Peer review” has a number of shortcomings.

There are two review processes: anonymous, in which the reviewers are not revealed to the authors (and sometimes the authors are not revealed to the reviewers), and transparent, in which both the reviewers and authors are known to each other.

We advocate transparency in the review process. Indeed, in Philippine Science Letters we encourage the reviewers to allow themselves to be identified.

In our opinion, anonymity in the review process means impunity, which could lead to abuse. Furthermore, anonymity conceals ignorance and incompetence.

We believe that a colleague or a collaborator can actually render a more constructive opinion on a manuscript than someone, who is in the same field, but who is a competitor. Indeed, it is common practice to ask a colleague to comment on a manuscript before it is sent off for publication. (It is also common practice to ask a colleague or collaborator to serve as reader or examiner of a student’s thesis.) We believe that we can always trust our friends and colleagues to give objective and helpful criticisms. We don’t think we can trust competitors to always do the same. 

Sadly, “peer review” results in “conformity,” i.e., if your views do not agree with those of the reviewers, your paper may not get published!

Some journals allow the authors to suggest possible reviewers. The rationale for this is the fact that the authors are probably the ones who know who are most familiar with the work they are writing about. Reviewers who are not actually working in the field may not be knowledgeable enough to provide a proper review.

The authors may also be allowed to name individuals who should not be asked to be reviewers. Those who are working in the same field are often competitors and may have “axes to grind,” or ulterior motives, which could cause them not to be objective in their assessment of the article. 

In our course, we strongly suggest that the work of competitors be properly cited (so as not to incite their ire). We counsel that it is probably best not to suggest reviewers who are known “enemies.” And we point out that reviewing of manuscripts, if done conscientiously, takes time and effort, so that to many, especially those who are active in research and other scholarly activities, reviewing is an imposition.

And then there are instances when a “peer” is too busy to review a submitted manuscript and would ask an underling (e.g., a graduate student, or a postdoc) to do the reviewing for them. In those instances, the review is not “peer review.”

(There are cases where authors may not bother with “peer review” and there is at least one journal that allows this: arXiv (http://arxiv.org/). Essentially, the authors are just letting the world know what they have done and are challenging others to prove them wrong.)

Journal decisions

Submitted manuscripts could be rejected, conditionally accepted after suggested revisions are made, or accepted outright for publication.

Rejection is not the end of the world. There are likely several “appropriate” journals for most articles. In our opinion, articles should be submitted, resubmitted, and resubmitted again, until it is finally accepted. Of course, if the reason for rejection is poor science, e.g., inadequate data, incorrect procedures, errors in analysis, etc., then major revisions would have to be made before resubmission. 

Revisions suggested by the editor or reviewers should be addressed properly. One could comply with the suggestions, or one could argue against them.

(Oftentimes, the reviewers’ constructive comments show the way toward improvement. A reviewer could suggest an experiment or two that one might have overlooked and could perform easily to complete the study. One could then resubmit the improved paper to the same journal, or submit it to another.)

Page proofs and copy editors

Once manuscripts are accepted for publication, they are usually edited and reformatted to conform to the style of the journal. The page proofs are usually sent to the authors for final inspection and correction. 

Most journals have copy editors who check the manuscript for typographical and grammatical errors, for incorrect wording and phrasing, and other errors in language. It should be recognized that copy editors do not necessarily know the particular science being reported, so that revisions made by them may not be appropriate and may result in scientific errors being introduced in the article.

The responsibility for ensuring the correctness and accuracy of the final version of the article lies in the authors.

At the End of the course, we ask the students to discuss a published article, or present a paper that they are in the process of writing, which is then critiqued.

* * *

Gisela P. Padilla-Concepcion is a professor at the Marine Science Institute, UP Diliman, and is currently UP vice president for academic affairs. She is a member of the NAST. She can be reached at [email protected]. 

Eduardo A. Padlan is a retired research scientist, formerly with the US National Institutes of Health. He is currently serving as an adjunct professor at the Marine Science Institute. He is a corresponding member of the NAST. He can be reached at [email protected].