Here’s a post by James Scott Bell at Kill Zone Blog: Stir your echoes
A writing echo is the close repetition of a word or phrase:
Monica charged into the room.
“So there you are!” she said.
Harvey said, “You don’t understand.”
The girl in the bed elbowed Harvey. “I think she does.”
“See you in court,” Monica said as she charged out the door.
The obvious echo here is charged. The words occur in close proximity. The echo clangs on the ear of the reader. It’s what I call one of those writing “speed bumps” that, even for a brief moment, can take the reader out of a smooth, fictional ride.
So don’t put them in.
But an echo is easy for a writer to write and overlook when editing his own manuscript. It should be something a good editor or reader catches for you.
Bell is SO RIGHT that this kind of repeated word is “easy for a writer to write.” He does not go far enough in his comment. Let me rephrase it more forcefully:
These echoes are AN ABSOLUTE PLAGUE UPON US.
Bell suggests doing a search to find words you tend to echo. Well, that’s a peachy idea, except that there is no specific word that I personally “tend to echo.” I mean, sure, maybe, but that isn’t the problem. If there were specific words, I could do a search for them as Bell says and there would be no problem.
But, no.
The problem is with every dratted word in the dictionary. It’s like the back of my brain says, OH! Let’s describe this guy as “sauntering!” And then for the rest of that page, the back of my brain continues to consider “saunter” and variations the ideal word for everyone moving anywhere. Then I’m over it and don’t use “saunter” or “sauntering” again during that book.
This is incredibly hard to spot when revising.
Please Feel Free to Share:
A script to catch exactly this probably exists somewhere. The problem would be finding it and deciding whether to pay for it.
Just from a quick search I see plenty of options for counting each word in a documen, which is a start at leastt; although for a manuscript it might take a while to run.
Mona, one problem is that you’d probably find dozens, maybe hundreds of examples where a specific word had been used three times in a manuscript — but it would be important to distinguish between three uses spread out over 420 pages versus three uses in two consecutive paragraphs. But! Maybe a computer person would find it trivial to set up a program that can tell the difference? I can’t tell what’s trivial from what’s impossible when it comes to things like that.
Ideally you’d also want to be able to tell your program, Ignore “said” and “was” and “is” and actually about a hundred other words. IDEALLY the person who put the program together in the first place would already have given it an extensive list of words to ignore.
Adding a condition like that would be pretty straightforward, I should think. You’d just define the range in which to search for a repeat, according to your tolerance (a paragraph—100 words? a page—500 words?). That should even make it run faster. And omitting common words is easy too, although it would have to be hard-coded as a list.
The difficult part is recognizing different forms of the same word— saunter, sauntered, sauntering (actually this kind of variation is easy); or worse something like fly, flew, flown (not sure this is possible).
Look, here’s a tool that recognizes frequent words and phrases among other things: https://expresso-app.org/
So it is possible. Unfortunately this one is web-based so you can’t run a doc through it. But the developer made the code available on github. If I were a programmer, I could take that and modify it for your purposes. Unfortunately I’m not that skilled.
Interesting, Mona. I will live in hope that someone produces something like that in the near future (and I find about about it).