The adverb prohibition was never a rule. It was a Strunk & White nervous tic that became orthodoxy — and our prose got worse for it.

The prohibition isn’t entirely wrong. There’s a real craft problem buried inside it, and the rule was an attempt to solve it. But somewhere between the original target and the modern reflex, the advice generalized from watch this one specific bad habit to cut the entire grammatical class — and most working writers now run a global find-and-delete against any word ending in -ly without ever asking which words actually deserve to go. The result is prose that’s lost its sense of time, its voice modulation, and its precision.

So today I’m going to do two things. First, separate the bad habit the rule was trying to fix from the grammatical class the rule swept up by accident. Second, give you a diagnostic move that’s better than the prohibition it replaces.

The rule started somewhere real. To make the case I want to make, I have to first concede the case the rule was making.

The problem is the she said sadly pattern. Manner adverbs grafted onto dialogue tags to do emotional work the dialogue itself failed to do.

“I don’t think we should see each other anymore,” she said sadly.

That line is doing two jobs and failing at both. The dialogue isn’t carrying the sadness — there’s no fragmentation, no hesitation, no syntactic evidence that the speaker is in pain. The adverb is being asked to label what the line should have rendered. And the label has a specific cost: it makes the writer visible, gesturing at the emotion instead of putting it in the prose.

Compare:

“I don’t think we should see each other anymore.” She didn’t look up from the cup she was holding.

The dialogue and the body language do the work. The reader gets there without being told. The sadly was redundant if the line worked, and a patch over weakness if it didn’t — and either way, it didn’t earn its place.

This is real ground. The dialogue-tag adverb is, most of the time, a writer covering for prose that didn’t do its job. Stephen King’s famous line — the road to hell is paved with adverbs — was diagnosing exactly this pattern, and his diagnosis was correct. The rule was earning its keep. But it was doing that work on one specific habit, in one specific structural position. And the move that came next is where the rule went off the rails.

The prohibition that got encoded into craft orthodoxy didn’t say watch your dialogue tags. It said: cut adverbs. All of them. On sight. The original target was manner adverbs after dialogue attribution — and the rule, somehow, generalized to an entire grammatical class.

There are at least five categories of adverb the rule swept up. Only one of them deserves the blanket treatment.

Manner adverbs in dialogue tags. The original target. She said sadly, he said angrily. Almost always patches over dialogue that should have done the work itself. Cut these.

Manner adverbs outside dialogue. Sometimes redundant, often irreplaceable. He turned slowly and He turned are not the same sentence. The slowness is doing emotional, temporal, and tonal work the verb alone can’t carry. Treating the second as automatically better than the first is a craft failure in the opposite direction — purifying the prose by stripping out the specificity that made it precise.

Temporal adverbs. Eventually. Finally. Soon. Later. These do pacing work that no substitute does as cleanly. He came back to the question is a different statement than Eventually, he came back to the question. The second carries the weight of a span the first doesn’t — and the only way to render that span without the adverb is to add three sentences of intervening time. The adverb is doing the work of three sentences in one word.

Sentence-modifying adverbs. Admittedly. Honestly. Frankly. Presumably. These do voice work, especially in first person. They mark a narrator’s relationship to their own statement. I should have noticed earlier is a fact. Admittedly, I should have noticed earlier is a fact filtered through a narrator who’s conceding something to the reader — and that filtering is the entire texture of the prose. Cut these and you flatten interiority. The narrator stops sounding like a person with a point of view and starts sounding like an annotated box of facts.

Intensifiers — the precise kind. Barely. Scarcely. Narrowly. Exactly. These are arithmetic. She barely finished is a different statement of fact than She finished. The first reports that finishing was contested. Sweeping these out costs precision the prose can’t recover by other means.

And then the lazy intensifiers. Very. Really. Quite. Somewhat. Yes — mostly cut these. The prohibition was right about this category. But it’s the smallest of the five, and the rule’s broad application punishes the other four for the sins of this one.

That’s the shape of the failure. A rule against one specific bad habit generalized into a rule against five distinct grammatical functions. We threw out a structural class to fix a stylistic tic.

So here’s what to actually do when you find an adverb in your draft.

Don’t delete it. The question isn’t should this go? The question is: what is this adverb compensating for? The adverb is a diagnostic marker — a symptom pointing to something else in the prose that may or may not be working. Your job is to figure out which. There are three possible answers.

Nothing. The adverb is doing its own specific work — temporal, sentence-modifying, precise intensifier, or a manner adverb carrying fine-grained character work no substitute would render. Keep it. He turned slowly stays. Eventually she came back to the question stays. You found the adverb, you ran the test, the adverb passed. Move on.

The verb is generic. This is the case the prohibition was right about beyond the dialogue tag. Walked quickly is a sentence asking to be hurried, rushed, strode. The adverb is patching a weak verb, and the fix is upstream, not downstream. Replace the verb with one that already carries the modifier inside it, and the adverb usually falls out on its own. The deletion-on-sight approach gets the right answer here — accidentally, and for the wrong reason. The reason isn’t adverbs bad. The reason is the verb wasn’t pulling its weight. And knowing the reason matters, because when you make this fix you’re not just cutting an adverb — you’re upgrading the verb. That’s where the actual improvement happens.

The dialogue is weak. The original case. “I don’t know,” she said hesitantly is a line where the dialogue itself failed to render hesitation. The fix isn’t a different attribution — it’s rewriting the line so the hesitation lands without needing a label. Add a beat. Fragment the sentence. “I don’t know.” A pause. “I really don’t.” If you can’t get the dialogue to carry the feeling, the dialogue is the problem. Not the adverb.

Three answers, three different fixes — and none of them is delete on sight. The discipline is the pause before the delete key.

One honest concession. There are writers whose prose genuinely improved when they policed adverbs harder. King is the canonical case — On Writing is partly a confession about his own habit and the work he did to break it. For some writers, especially early on, especially in first drafts, the blanket prohibition is useful training wheels for a real problem.

But the question isn’t whether the rule was ever useful. It’s whether you’ve graduated from needing it. Most working writers have — and most don’t notice the rule has stopped earning its keep and started costing them precision, because it operates as a reflex now, not as a question. The honest move is noticing when you’re using a rule as a substitute for thinking. The diagnostic takes more thought than the prohibition does. That’s a feature, not a bug.

The prohibition was always a fix for one specific bad habit. We extended it to a grammatical class, and the prose got blander while the writers who promulgated the rule got famous for breaking it. Read King carefully and you’ll find adverbs everywhere his prose is doing serious work — temporal adverbs holding decades together, sentence modifiers filtering a narrator’s voice, precise intensifiers measuring a moment’s stakes. He never followed his own rule. And he was right not to.

So the next time you find an adverb in your draft, don’t reach for the delete key. Ask what it’s compensating for. Sometimes the answer is a stronger verb. Sometimes it’s a rewritten line of dialogue. And sometimes — more often than the orthodoxy admits — it’s nothing at all, and the adverb stays. Because it was never the problem.What you call a thing is never neutral. It’s the smallest unit of theme your book has. Use it.

Until next time, thanks for listening.