Physicists on Their Own When Learning to Write
“As a scientist, you are a professional writer,” asserts Joshua Schimel in Writing Science, one of my favorite science-writing guidebooks. Indeed, writing sits at the heart of every researcher’s livelihood, as the quality of their papers and grant proposals can profoundly shape their scientific careers. But what training do physics students actually receive to develop their writing skills? Briley Lewis of the University of California, Santa Barbara, set out to answer that question through a large-scale survey of hundreds of physics and astronomy students across the United States [1].
The results paint a sobering picture of the situation of writing education—one that only worsens as students advance to grad school. While undergraduates pick up some skills through general writing courses or lab-based assignments, structured instruction nearly disappears in graduate school. For nine out of ten respondents, learning to write as a scientist is an improvised process—shaped by self-teaching, piecemeal on-the-job experience, and whatever feedback a supervisor happens to provide.
Lewis approached this problem from a unique perspective—that of a professional astronomer with a passion for creative writing and some freelancing experience. Those interests led her to pursue a certificate in writing pedagogy alongside her PhD in astronomy and astrophysics at UCLA. Now a postdoc supported by the National Science Foundation, she has dedicated the “broader impacts” component of her fellowship to investigating how physicists learn to write and what might be done to support their development.
Lewis first set out to rigorously document the state of writing education. “I had a hunch from my own experiences, but I wanted to see if that anecdotal experience tracked with other people’s experiences,” she says. She launched an extensive survey, recruiting participants through professional-society mailing lists, departmental forums, conference flyers, and word of mouth. The respondents were 515 physicists and astronomers at the junior-year undergraduate level and above, who had completed at least part of their training in the United States. The sample spanned more than 200 institutions nationwide, from community colleges to Ivy League research universities.
Nearly all the respondents recognized writing as an essential skill, but few of them have received structured instruction in scientific writing. Instead, the primary form of writing education for 62% of participants was mentor feedback. Graduate-level electives specific to physics and astronomy were available to only 11% of respondents—and even when offered, many students skipped them, deterred by heavy workloads or by the sense that such courses aren’t necessary. One out of six of the survey’s participants reported receiving no scientific-writing training at all.
These results aren’t particularly surprising, says Andi Piña, a physics-education researcher with the Rochester Institute of Technology in New York. “However, to have those notions backed by data is powerful, and I hope it pushes educators to reconsider how we integrate writing into physics curricula, particularly in graduate education,” Piña says. Suzanne White Brahmia, a physics-education researcher from the University of Washington, agrees that academic programs should include structured scientific-writing education. “Funding agencies should support the developmental work necessary to make this happen,” she says.
Mentor feedback might seem like the most natural form of apprenticeship—passing skills from experienced scientists to their trainees. But Lewis cautions that this approach is fraught with problems. First, it is highly inconsistent: Some mentors are thoughtful and supportive, while others are neglectful or outright dismissive. Second, the “power-infused” relationship between mentor and student adds a layer of difficulty. “The success of your career and your project really depends on this person, and there is a lot of potential for that to go wrong,” Lewis says. Finally, mentor guidance often occurs in high-stakes situations, such as preparing a publication for peer review. Students may miss out on the lower-stakes practice that promotes real learning.
Many respondents described positive experiences, praising supportive mentors and acknowledging institutions that promoted writing training. But some of the reports were negative, recounting experiences that Lewis found concerning. One participant wrote: “My educational path completely failed to prepare me for a career that lives or dies on the ability to write well. What I now know is essentially all self-taught.” Another recalled mentor feedback as “Awful. Absolutely awful. I cried during the meetings.”
What can be done to better prepare scientists? Lewis has outlined a few possible paths forward, “none of which are admittedly easy.” They include training mentors and making writing a required—rather than elective—part of curricula. “These are pretty big changes,” she acknowledges, wary of adding to the heavy workloads of students and professors. “But I just want more people to think about [these issues].” She plans to tackle one recommendation herself by creating online resources accessible to anyone at any institution.
Lewis’s survey didn’t address the impact of artificial intelligence (AI) and large language models, as most of the respondents were educated before these tools became available. White Brahmia thinks that AI tools can have a positive impact. “When used responsibly, AI can help lower barriers to persistence in the sciences, particularly for non-native English speakers,” she says. However, both Lewis and Piña believe that AI still produces poor writing and that its use can, in many cases, do more harm than good. “Relinquishing the task of writing to AI would take away from the learning that happens in the writing process,” says Lewis. The advent of AI means that “it is now more urgent than ever to provide some structure to writing education in STEM,” says Piña.
–Matteo Rini
Matteo Rini is the Editor of Physics Magazine.
References
- B. L. Lewis, “Surveying the state of writing education in physics and astronomy,” Phys. Rev. Phys. Educ. Res. 21, 020146 (2025).