Active vs passive voice in the methods section: what journals want.

No major journal requires the methods section to be written entirely in the passive voice. The APA Publication Manual (7th edition, Section 4.13) tells authors to use the active voice as much as possible, and Nature journals state outright that they prefer authors to write in the active voice. Passive still belongs in methods, but only where the thing acted on matters more than the person acting; for every sentence that reports a decision, name the agent and use an active verb.

Where did the all-passive rule come from?

It is a leftover convention, not a current requirement. For much of the twentieth century, journal prose treated the passive as the sound of objectivity: the experiment appeared to run itself, and the researcher politely vanished from the page.

Journals moved on. The advice did not. It survives in writing courses, supervisor feedback, and translated style guides, and it lands hardest on researchers working in English as an additional language, because it arrives as a simple, checkable instruction: never write we. Simple rules travel well. This one is out of date.

What do style guides and journals say now?

The two most-cited authorities both prefer active voice and both leave room for passive. APA's grammar guidance says to "use the active voice as much as possible to create direct, clear, and concise sentences, especially when you are writing about the actions of people," and to use the passive "when it is more important to focus on the recipient of an action than on who performed the action, such as when describing an experimental setup." Nature Portfolio's author guidance is shorter: "Nature journals prefer authors to write in the active voice," with we performed the experiment as its own example.

Both positions share a shape. Neither bans the passive. Both make it the marked choice, the construction that needs a reason.

When is passive voice the right call in methods?

Use passive when the actor is obvious, interchangeable, or irrelevant to replication, and the material is the true topic of the sentence. This is exactly the carve-out APA names: the experimental setup.

Samples were centrifuged at 3,000 g for 10 minutes. Which technician pressed the button changes nothing about whether the result replicates, so the sentence rightly spends its subject position on the samples. The same logic covers Participants were randomly assigned to one of three conditions, where software did the assigning and naming it adds noise.

Passive also keeps a paragraph on topic. When four consecutive sentences all concern the same tissue sample, keeping the sample as the grammatical subject reads better than opening every sentence with we. Topic continuity is a legitimate reason to choose the passive. Habit is not.

When does passive voice hide who did what?

Passive hurts when the agent made a judgment call, because deleting the agent deletes information a replicator needs. Procedures can go agentless. Decisions cannot.

Outliers were removed. By what criterion, set before or after seeing the data, by whom? The diagnosis was confirmed. By the treating physician or by an independent rater? That single difference decides whether the ratings were independent, and a careful reviewer will ask. Compare: We excluded participants whose reaction times fell more than 3 SD from their condition mean. Every element of that sentence can be checked and repeated.

Agentless prose also breeds dangling modifiers. After incubating overnight, the supernatant was collected claims the supernatant did the incubating. Sentences like these are part of what reviewers gesture at with the vague complaint that the English needs improvement, and a wall of agentless constructions is one of the recognized signs a paper reads like a translation, since many languages, Arabic among them, use an agent-suppressing passive far more freely than modern journal English does.

There is a cost in words, too. Centrifugation of the samples was performed spends seven words doing the work of four, the same inflation pattern behind most of the wordy phrases journals keep cutting.

How do you mix voices so a paragraph reads naturally?

Let the agent decide, sentence by sentence. Decisions take we and an active verb. Standard procedures and materials take the passive. The paragraph alternates, and no reader notices.

We recruited 142 undergraduates through the university participant pool. Participants were tested individually in a sound-attenuated booth. We excluded four for failure to follow instructions, leaving 138 in the final sample. Stimuli were presented in random order on a 24-inch monitor.

Two decisions in active voice, two procedures in passive. Neither an APA copyeditor nor a Nature editor would flag a word of it. The test is short enough to apply mid-draft: if the sentence answers "what did you decide?", write it active; if it answers "what happened to the material?", passive is fine.

How does a rule-based check tell acceptable passive from overuse?

By treating passive as a clarity question rather than an error. Waraq's built-in rule library words it exactly that way: the rule grammar.passive.overuse, titled "Passive voice," reads in full, "Prefer active voice where it improves clarity." The id carries the point in its last word. Overuse is the problem, not existence, so when a manuscript review applies the rule, a setup sentence like the centrifugation example has nothing to answer for, while a string of agentless decision sentences does.

You can run the same check without any software. Go through your methods section and mark every passive sentence that reports a choice you made: exclusions, thresholds, model selection, stopping rules. Convert those to active, leave the procedures alone, and read the section once aloud. What remains is a methods section in both voices, each one earning its place.