The Ethics of the Pretty Sentence

An editor once improved a line of mine, a testimony from a man named Pascual describing his brother's funeral, from "there was no priest, so we buried him ourselves, and it took longer than it should have" to "in the absence of clergy, the brothers laid their kin to rest themselves, a slow and hallowed labour." I did not thank her. I asked her to put it back.

The second version is, by any workshop standard, the finer sentence. It has rhythm, a rise at the end, a deliberate word choice in "clergy" instead of "priest." It is also no longer describing what happened to Pascual. It is describing what happens in a certain kind of nonfiction prose when grief is the subject, and that kind of prose has its own gravitational pull, toward solemnity, toward rhythm, toward language that performs respect rather than earning it by staying accurate to what a specific man actually said about a specific, badly organised afternoon.

The beautification tax

I call the gap between what a speaker said and what an editor's improved version says the beautification tax, and I calculate it the same way I calculate the register ratio in other work, except here the question is not which words survive but which facts survive the upgrade. Pascual's sentence, in its original form, contains three facts: no priest attended, the brothers did the burial themselves, and it took longer than expected, which in context meant longer than dignity would have preferred, a small, specific admission of disorder. The improved version keeps two of the three facts and replaces the third, the admission of disorder, with an assertion of solemnity that Pascual never made. He did not say the labour was hallowed. He said it took too long. The tax on that sentence is one full fact, quietly billed to the reader in exchange for a smoother rhythm.

Why the fee is not obvious

What makes the beautification tax dangerous rather than merely annoying is that it is invisible to a reader who was not there, and often invisible to the editor levying it, since the improved sentence does not read as false. It reads as more literary, and a certain kind of readerly trust attaches automatically to literary confidence: a sentence that sounds like it knows what it is saying about grief gets believed more readily than one that sounds like an actual person losing his composure over logistics. The tax collects precisely because the improved version is more persuasive, not less, which is the opposite of how a factual error usually behaves.

I went back through a manuscript I had allowed a different editor to polish early in my career and counted the tax across forty testimonies. Sixteen sentences had been upgraded in ways that added a fact the speaker never stated, almost always in the direction of dignity, coherence, or acceptance, virtues the editor clearly valued and clearly assumed the speakers would have wanted attributed to them. Not one upgrade added disorder, anger, or confusion, the actual textures of Pascual's afternoon and of most of the other accounts in that book.

The honest objection

You could argue that some smoothing is simply craft, that no transcript survives verbatim into print without judgment calls, and that calling every improvement a tax risks treating all editing as theft, which would make the job impossible. I agree that all editing involves judgment and I am not arguing against judgment. I am arguing against a specific, recurring pattern in that judgment, the tendency to upgrade toward solemnity whenever the subject is loss, because solemnity reads as respectful and disorder reads as unpolished, when in Pascual's case the disorder was the actual content of his grief and the solemnity was a costume put on it after the fact, by someone who was not burying her brother that afternoon and never had to.

What I do instead

I still edit for clarity, and I still cut repetition and false starts that a spoken account accumulates and a printed page does not need. What I no longer permit myself is the specific move of adding a fact the speaker did not state in service of a more literary register, and I have started reading every edited testimony backward against the recording, sentence by sentence, asking only one question of each change: did this remove something the speaker said, or add something the speaker did not. Cuts I allow freely. Additions, even beautiful ones, get sent back, because the fee for beauty in this genre is always paid by someone who is not in the room to object.