Scientific writing 3.0: A reader and writer's guide - Jean-Luc Lebrun, Justin Lebrun 2021

Abstract versus conclusions
Conclusions: the smile of your paper
Paper structure and purpose

After ruling out many choices, I decided that no part of the body could represent the conclusions better than the smile. Why a smile? I thought again of the many conclusions that had disappointed me, and deflated my enthusiasm with self-deprecatory endings such as ’In order to test real performance improvement…,’ or ’This could be greatly improved by…’. I had read these articles with great interest until, in the conclusions section, I had found suggestions that nothing significant had been accomplished. I felt like the person about to buy a car described as safe, only to be told at the last minute that the car had no air bags and no antilock braking system. Previously undisclosed limitations disguised as future work frequently surface in the conclusions to disappoint the reader who genuinely assumed the author had dealt with them already. Imagine a lawyer who managed to demonstrate the innocence of his client throughout the court proceedings, but who, on the very last day in front of the jury, apologizes because not enough evidence has been produced to justify the plea of innocence. How unbelievable!

The way a defense lawyer ends his plea in front of a jury should also be adopted to end a scientific paper: with assurance, firmly, and smiling, trusting the jury will find the client not guilty of scientific insignificance. Lawyers know that the day the jury gathers for the final plea is an important day. The day the writer writes the conclusion of his paper is also an important day. The writer cannot write it at nighttime, close to exhaustion. His writing will be lifeless. The writer cannot write the conclusion too long after the end of the research. His sense of past achievements may be gone.

So, before writing the conclusion, the writer has to re-energize his pen by reading again introduction and discussion to identify the research milestones. He has to accumulate the intermediary scientific merit points to form his final score: the paper’s global contribution. Re-energize yourself, smile as you consider your score, stay positively charged… because the negative ions are there: your fatigue, the time that passed since the end of your research, and the limitations that need to be addressed in the future. However, do not waste that positive energy on yourself, to polish your halo, or bask in the sunshine of your glorious past. The conclusion is not an opportunity for an ego trip. It is an opportunity to polish, not your halo, but your diamond of a contribution because you need to sell it to cash in on citations, to encourage others to use your work.

You may have noticed that, in some journals, articles have no conclusions heading; the discussion ends the paper. The need to conclude is still there however, even if the heading is absent. Some journals — Nature is one of them — recommend to finish an article without conclusions. They would rather have the author write the last paragraph “about the implications of what the reader has read”,¹ and not summarize what has been accomplished. Professor Railsback with great common sense writes: “Conclusions are just that — the inferences that can be drawn from your data, not a reprise of the entire paper.” Combining expert advice, ’inferences’ and ’implications,’ it becomes clear that conclusions are not just another abstract.

Abstract versus conclusions

Surely readers wouldn’t notice that your abstract is similar to the conclusions, would they? They would!

Readers read in a non-linear fashion. They tend to skip large sections of a paper, jumping from abstract to conclusions for example, like the hurried reporter only attending the first and last day in court. From a writer’s perspective, that behavior is not ideal, but the writer can use this knowledge to his or her advantage. First, the writer now understands how dangerous it is to copy and paste sentences between abstract and conclusions since the reader immediately notices them. Second, the writer should differentiate the conclusions from the abstract to avoid boring the reader. How do the two differ?

· Sometimes, the journal recommends the use of the past tense in the abstract. Unfortunately, the main tense used in the conclusions is also the past tense because you are referring to what you did. Only the facts that have been demonstrated without a doubt, the unquestionable scientific facts, are stated in the present tense. The lawyer says ’my client is innocent,’ not ’my client was innocent’. The present tense in the conclusions reinforces your contribution. If the journal does not impose the use of the past tense in the abstract, it becomes advantageous to write the whole abstract in the present tense because doing so differentiates conclusions from abstract.

· Whereas the abstract briefly mentions the impact of the contribution, the conclusions dwell on this aspect to energize the reader. In his book A Ph.D. Is Not Enough, Professor Feibelman gives his writer’s viewpoint.

“The goal of the conclusions section is to leave your reader thinking about how your work affects his own research plans. Good science opens new doors.”²

· Conclusions are more comprehensive than the selective abstract. Conclusions bring closure, not on what the self-contained abstract announced, but on what the introduction and discussions opened. They close the door on the past before they open ’new doors’ onto the future.

· The abstract adopts a factual, neutral tone. The conclusions keep the reader in a positive state of mind. Remember that a reader needs strong motivations to read the whole paper, not just your conclusions. The motivating role is traditionally taken by the introduction, but if the reader skips it and jumps directly to the conclusions, then they must also motivate the reader to reach inside your paper. Therefore, keep your energy level high and think positively about your contribution.

· Everything in an abstract is new to the reader. In the conclusions, nothing is. The conclusions do not surprise the reader who has read the rest of your paper. Following the analogy with the defense lawyer’s final plea in front of a jury, it stands to reason that any attempt to convince the jury at the last minute with evidence that has not been cross-examined is not receivable and is objectionable. Such last-minute-theatrical surprises are the realm of Hollywood movies only. Even the section about future work is expected. In the discussion section, you venture explanations that require future validation, or you suggest that different methods might be helpful to bypass constraining limitations. The reader who has read your discussion anticipates that, in your future work, you will explore these new hypotheses or use these different methods.