Four kinds of writing “errors”and how you can effectively respond when you spot them

“Oops” is only one of many possible responses when writing isn’t working. And in most cases, “oops” is NOT the most helpful framing. (Image by Samiran Modak from Pixabay)

Dear Troublemakers,

When you’re working with a developing writer, you may be doing a lot of different things – but one thing you’re almost certainly doing is reading drafts and making suggestions for improvement. My co-author Steve and I recently wrote about how that draft-reading work can (should!) shift as the writer develops and as any particular draft develops.

Today, though, we’re looking at another angle. Whether or not you’ve had any training in mentoring earlier-career writers (and most academics haven’t – hence our book!), there’s one thing you surely can do. That is: look at a draft manuscript and flag things you think ought to be revised – “errors”, if you will.

The developing writers you (and we) work with are trying to master a new writing genre – scientific writing – with its attendant conventions of format, style, and content. A lot of the things you might flag in a manuscript are related to students’ unfamiliarity with these conventions. (These aren’t the only issues, of course, but we’ll lay more fundamental writing problems aside for now.) You’re likely familiar with scientific-writing conventions, and you’re probably rather good at spotting things that appear not to conform with them.

There’s an important caution here, though: because you know scientific writing has conventions, you may feel the need to constrain writing to very close observance of these conventions. This impulse can be manifested in a desire for writing to sound science-y, and that desire is a big part of how our literature came to be the turgid, tedious mess that it is! But scientific writing has more latitude in style than either developing writers or their mentors often realize. [1] Essentially, you’ll give more effective feedback if you keep in mind that scientific writing is a specific type of writing with many conventions, but not nearly as many rules as we’ve been led to believe.

With this context in mind, imagine that you’re reviewing a developing writer’s work and a chunk of text – a word, a phrase, a paragraph – jumps out at you as needing correction or improvement. Before writing that comment in the margin, we suggest asking yourself where that chunk sits in the following categories. 👇

Quick-view version of the descriptive text below. Click on table image to view a PDF (This blog platform does not support tables. :/)

(1) Does the chunk of text do something that’s inconsistent with good writing in general? For example, have you spotted a spelling error, or a grammatical one, or a paragraph that lacks a topic sentence?

(2) Does the chunk do something that’s perfectly fine in other genres, but violates reader expectations for scientific writing? For example, does it depart from conventional IMRaD [2] structure, or does it use multiple direct quotes from cited material? [3]

(3) Does the chunk do something that’s common in our literature – even typical of it – but that we’d be better off abandoning? For example, is it copiously spattered with acronyms, is it written exclusively in the passive voice, or are its sentences long and complex?

(4) Does the chunk do something that’s common and acceptable in our literature, but that you tend not to do yourself? Most writers (academic or not) have preferences and pet peeves, and many of us are tempted to elevate our preferences to rules – but we shouldn’t. Perhaps the student you’re mentoring prefers “crucial” to “critical,” while you’re the other way around? Or perhaps the student included a summary of major results at the end of the Introduction, while you, a murder-mystery fan, hate to give away the ending?

So how does categorizing the chunk of text this way help? Well, we hope it’s obvious that these four kinds of writing “errors” (and now you understand the scare quotes) should be handled differently. A type 1 issue really is an error4 to be fixed, and you can refer the student to a general guide to English composition. A type 2 issue is similar, but it can provoke a useful discussion about our genre conventions for scientific writing (for example, how we came to adopt the IMRaD format, and why it’s useful that we keep doing so). Here, you can lean on a more specialized guide to scientific writing in particular. A type 3 issue is a great place to kick off a discussion of what our literature is like vs. what we’d like it to be like, and of how and why we need to consider the reader as we draft and revise our writing.

Finally, type 4 “errors”. A type 4 “error” (and now the scare quotes have really come into their own) is a great place to just back off. We all have preferences; but we sometimes need to recognize that that’s all they are! Type 4 issues (we’ll stop calling them “errors” now, even in scare quotes, because they simply aren’t) are your opportunity to help a student make their own choices about style and voice. The more practice they have at that, the sooner they’re likely to grow into independence as an academic writer.

In other words…👇

Quick-view version of the error types with corresponding effective responses. Click on the table image to view a PDF version (This blog platform does not support tables. :/)

The key idea here is that you can be more effective and efficient if you calibrate your feedback to the type of revision the student makes. But there’s a bigger point here, too. This isn’t just about commenting on our mentees’ writing.

One frustrating challenge we each face (no matter what our career stage) is this: writing effectively while conforming to reader expectations for our genre, but also while retaining our own voice.

If you give too much deference to convention and too little thought about what’s “wrong” vs. what’s a style choice, you are asking a student to suppress (or never develop) their own writing voice.

Above all, not being able to bring their own writer’s voice to the page leads a student to conclude that they can not bring their whole self to science. This is profoundly problematic, given the history of science excluding most people’s knowledge, perspectives, and voices. And almost certainly if you’re reading our work, you welcome students and colleagues bringing their whole selves to the page! So, we hope you’ll find these guidelines in today’s post useful as you work to ensure developing writers do feel welcome in science.

Beyond that vital, global point, there’s another key one you’ll want to keep in mind. Without a writer’s voice, scientific writing becomes soulless and tedious. And tedious writing costs: it makes it harder for our readers (even for us as we write) to get excited about the science we’re writing about. It makes it harder for readers to understand what we’ve done and learn from it or replicate it. There’s even evidence that readers are more likely to resist or stop engaging with writing that’s harder, or less interesting, for them to read.

Ultimately, we are confident you can help developing writers balance genre conventions with their own style and voice. Recognizing and discussing these four kinds of writing “errors” can take you and the writers you mentor a long way towards that goal.

How about you?

What types of errors do you usually see in the writing you mentor? And, which types of “errors” do you find it hardest to respond to effectively?

© Stephen Heard and Bethann Garramon Merkle March 10, 2026. This post is based on material from our new book, Teaching and Mentoring Writers in the Sciences – especially Chapter 5.

---

NOTES

[1] It’s OK, for example, for scientific writing to be beautiful, or for a title to be funny – always respecting the crucial importance of writing that’s clear. If you’d like to enjoy an extended metaphor about when and where to embellish writing, check out the concluding chapter of The Scientist’s Guide to Writing. Early in the development of scientific writing as a genre, the chemist and polymath Robert Boyle suggested that it’s actually a good idea to decorate text with enjoyable or beautiful bits (painting a telescope’s tube); but that it’s not a good idea to let that decoration interfere with the text’s usefulness (painting a telescope’s lenses). It’s been 365 years since Boyle made this point, but if anything, the modern explosion in the size of our literature has only made it more relevant.

[2] IMRaD is an acronym for a common research article structure: Introduction, Methods, Results, and Discussion.

[3] Direct quotes are rarely used in scientific writing, but are extremely common in academic writing from disciplines such as the humanities. One reason (of many) for this difference is that the actual words are held to matter in humanities discourse (not merely the content), while in science it’s primarily the content that’s held to matter. (It’s interesting to consider how often this is actually true, vs. being a cultural difference between fields. If you want to nerd out on research relevant to this, science studies is probably your best bet.)

[4] Well, except that there are plenty of things many writers think are grammatical errors that simply aren’t – like split infinitives. That’s a bigger topic than a footnote has room for; we’d just encourage you not to rely completely on your instinct for what’s really a “rule” in English and what isn’t. Selecting a specific style guide and standardizing its use in your research group and in the courses you teach can help your students (and you!) start identifying what’s a style choice and what’s a grammatical error.