Proofread by Ears: Dictation for Copy Editing

Ask any copy desk which publishing role eats the hands fastest and the answer is unanimous: the line editor, which is why dictation for copy editing is the quietest change in the trade. The line edit sits at the spillover of the manuscript, sentence by sentence, and the work is constant, dense, and keyboard-bound: delete a phrase, re-type a clause, cut a comma, move a paragraph, fix a reference, and do that for a chapter, then a book. It is not the writing volume of an author and not the browsing of a proofreader; it is more edits per hour than any other literary work, and every edit is a burst of typed micro-surgery on text someone else wrote. The wrists pay for the permanence of the profession.

The peculiar cruelty of line editing is that the high-edit work is also the highest-concentration work. An editor reads at the top of their attention, then types corrections at the same level, and both acts demand the same hours from the same pair of hands. The manuscript does not care; it simply arrives in chunky files and expects full attention. Meanwhile the body's needs, the pause, the stretch, the lowering of the wrists, are all politely deferred until the chapter ends, and the chapter rarely ends before the forearm does. Editors with twenty years in the trade can usually date the beginning of their hand issues to a particular marathon project.

The freeing realization is that most of the line edit is not typing; it is decision. The editor decides the sentence is clumsy, decides what it should say, decides where the comma belongs. The typing is only the act of recording the decision, and recording is what the voice does well. Dictation edits a line by saying the replacement aloud, which is exactly how an editor already evaluates rhythm in their head, and the hands are then reserved for the movements that genuinely require them: the drag, the select, the delete, the insertion of a reference. The strain leaves the repetitive pair and stays only in the precise moments.

An editor's corrections are a punctuation-rich language, and dictation handles them perfectly once the shorthand is settled. The core commands are the ones a line editor uses constantly: full stop, comma, new paragraph, and the family of clause commands, with colon, semicolon, and dash spoken as themselves. Around them sits the editorial vocabulary: 'delete that,' 'replace with', 'move this up', 'swap the order', 'underline it', 'insert a footnote', 'capitalize it.' Each phrase is short, unambiguous, and spoken at the moment the editor's eye lands on the problem, which is exactly the rhythm the edit requires. The hand never leaves the manuscript's cursor.

The trick that makes the shorthand reliable is to keep it minimal and constant. Editors who build a private phrase vocabulary, and use the same ten or fifteen phrases every project, find the engine's accuracy climbs quickly, because repeated phrases are learned phrases. A second trick of equal value is to state the target before the instruction: 'in the second paragraph, replace so with thus.' The target-first order keeps the edit precise and gets the correction into the right place even when the editor's eye is moving fast down the page, and it also produces a transcript that reads like a sensible edit log.

Resist the temptation to dictate sub-sentence surgery on complex structures until the vocabulary is fluent. Begin with the replacements and the deletions, the highest-value, easiest commands, and add the structural moves in the second week. The editor who encodes the shorthand gains a second superpower with it: the corrections become auditable, as the spoken edit log is a draft of the editor's notes, and the commentary on why a sentence needed changing, which was always half hidden in the margin, now exists as a plain narrative that can be shared with the author and the team. Fluency first, ambition second.

The read-aloud pass is the oldest trick in the editor's craft, and it is the one skill that dictation turns into a routine rather than an indulgence. The mechanics: the manuscript is selected, the software reads it back, or the editor reads it aloud in their own voice, and the ear follows the rhythm while the eye is freed from the line. What the ear catches is a different class of error than what the eye catches: the repeated word two sentences apart, the clause with its subject missing, the sentence that promised one structure and delivered another, the over-long wrap of a parenthesis, the wrong referent of a 'this.' The eye reads what is written; the ear reads what was meant.

The pass happens in two directions. First, the voice reads the manuscript aloud, and the editor types the corrections flagged by the ear. Second, and this is the deeper use, the editor dictates the fix in the same breath, saying the better line before the memory of the stumble cools. The two acts merge into a single workflow: hear the fault, speak the fix, move on. Editors who run this report a change in the type of fatigue as much as the amount: the eyes tire slower, the chair starts feel less like a commitment, and the end-of-day headache that used to trail a long edit gives way to a different, more precise tiredness that belongs to the concentration, not the posture.

The read-aloud pass is also the fastest quality check on the voice's own work. A dictated correction that reads back clean, with the ear confirming the flow, is a correction the editor trusts. A dictated line that stumbles on its own rhythm is the signal to look again, and the editor gets a free re-read of every change they have made. For the long view, the read-aloud habit is what turns dictation for copy editing from a tool into a craft: the ear becomes a permanent second reviewer, and the manuscript comes out of the desk with the fingerprints of two senses rather than one.

The serious experiment is a 40,000-word manuscript: a novella, a slim nonfiction title, a long report. Choose a manuscript you would edit anyway, for whatever reason, and run the first half with the classic method, keyboard and screen, then the second half voice-first, dictating the corrections, with the same style sheet and the same quality bar. Log the daily hours, the end-of-day hand score, and your own sense of the edit's depth: how many pass-two corrections each half still needed. The comparison is unfair in the best way, because it is your own body and your own standards on both sides.

The first thing the test reveals is speed, and it is not what editors expect. Editing is not writing; nothing in the craft is faster than precision, and a dictated correction is not automatically faster than a typed one. The early chapters of the voice half run comparably, then slightly ahead, because the read-aloud pass becomes part of the edit rather than a separate ritual, and the second pass shrinks: the weary re-reads that the eye used to need are replaced by the ear's continuous surveillance. The whole 40,000 words typically lands in fewer hours, with the difference concentrated in the later sessions, when the old method slows with fatigue and the voice method does not.

The second revelation is the hand score, and it moves as expected: the second half of the manuscript costs the editor a fraction of the evening tension the first half did. The third finding is the one that stays with editors: the miss rate. When they compare the two halves' final errata, the voice-first half does not embarrass itself; it holds its own, and it produces a distinct, complementary profile of misses, the homophone and the mis-captured clause, which the typed half never produced and which the read-aloud pass catches on its own. The copy desk test's verdict, after one manuscript, is typically a switch.

The long-term habit is where the craft develops, and the first wisdom is to keep the modes differentiated rather than blended. Dictate the edits that change sentences, and type the micro-moves: the reference insertion, the character deletion, the formatting. An editor who tries to speak every action loses the precision that the typing hand was always best at, and an editor who types every sentence loses the strain relief that brought them to the voice. The split is not a compromise between methods; it is the assignment of each job to the instrument that does it best, and the ratio settles over a season into something like three spoken edits for every typed action.

The second wisdom is to let the habit spread upstream into the other editorial duties where it never expected to go: the query notes to the author get dictated and cleaned, the editorial letter is spoken first and then polished, the style sheet updates are narrated into the standards file, and the fact-check calls are recorded as text while the phone conversation runs. Every one of these is prose with a listener, and every one of them used to be typed from scratch. The voice becomes the editor's default for all first-draft text of any kind, and the keyboard becomes the editor's instrument for the corrections.

The final wisdom is the professional one: an editor who edits by ear becomes a better editor, because the two senses do different work. The eye sees the line; the ear hears the paragraph. The combination is the slow, permanent improvement in the trade itself, the ability to catch the rhythm break that neither alone would reliably find. And the hands, which used to be the profession's quiet tax, simply stop being a topic. The editor retires with the same reflexes they came in with, and the wrists leave work on the same page they started on. That, more than any word count, is the argument for the habit.