October 1, 2026
Writing Is Part of the Science
I recently came across an interesting article in Nature Biotechnology titled “An open letter to graduate students and other procrastinators: it’s time to write”, by Dennis J. Hazelett.
The article begins with a problem that is probably familiar to almost anyone who has done research: understanding the structure of a scientific paper is relatively easy, but sitting down to write your own work can feel surprisingly difficult.
There is a good reason for that.
When we write about our own research, we are not simply describing something that already exists in a textbook. Ideally, we are trying to explain something new — a question that has not been answered before, an experiment that has not been performed in exactly that way, or an interpretation that has not yet been presented.
That makes scientific writing fundamentally different from simply reporting information.
One point in the article particularly caught my attention. Before worrying about all the details of a manuscript, a researcher should be able to answer two basic questions:
What is the problem or question I am trying to address?
and
What do I actually think about it?
These sound like simple questions, but they are surprisingly powerful.
During research it is very easy to become absorbed in experiments, measurements, simulations, troubleshooting, data processing and endless technical details. All of these activities are necessary. But at some point we have to step back and ask what all of those results actually mean.
Writing forces us to do that.
It forces us to distinguish an interesting result from an irrelevant one. It exposes gaps in our reasoning. It makes us decide which evidence really supports our argument and which observations are merely incidental. Sometimes we do not fully understand our own research until we try to explain it clearly.
This is why I think scientific writing should not be considered something that happens after research.
It is part of the research itself.
A laboratory notebook contains observations. A computer contains data. An experimental setup may demonstrate a phenomenon. But none of these automatically becomes scientific knowledge.
The researcher still has to construct the intellectual connection between the question, the evidence and the conclusion.
And that connection is largely built through writing.
Hazelett makes the argument even more strongly: the long-term influence of scientists depends not merely on having ideas, but on developing those ideas, supporting them with evidence and communicating them so that they can influence the work of others.
I think this distinction is important, particularly for graduate students.
It is possible to spend years becoming extremely competent at operating equipment, writing code, performing measurements or solving technical problems while continuously postponing the process of writing. Writing then becomes a large and intimidating final task.
Perhaps it is better to reverse that relationship.
Write while thinking.
Write while experimenting.
Write when a result does not make sense.
Write down why you believe an experiment should work before you perform it.
And after obtaining the result, write down whether it changed what you originally thought.
In this sense, writing is not simply the production of papers.
It is a method of thinking.
Good scientific writing ultimately requires clarity of thought. If we cannot explain what question we are asking, why it matters, what evidence we have collected and what we think that evidence means, the problem may not be the writing itself.
The problem may be that our thinking is not yet finished.
Source
Dennis J. Hazelett, “An open letter to graduate students and other procrastinators: it’s time to write,” Nature Biotechnology, 43, 447–450 (2025).