Wow, I had no idea how many times I used the word “very.”
Now, of course, I don’t mind using adverbs, in reasonable moderation. Certainly more than some people use. And “very” is exactly the right word sometimes, because people do use the word when they’re talking, right? Sometimes it just sounds right in dialogue. Plus you get phrases like “the very edge of the cliff”, where “very” is clearly irreplaceable. But, still, I think I removed over half of all instances of the word.
Only about a tenth of the semicolons, though. I actually thought most of those were good where they were. And I didn’t take out more than about a fifth of all the dashes. Especially the parenthetical dashes. I nearly always dislike parentheses in fiction, and a lot of the time parenthetical commas don’t work right. Hence the dashes. Plus how else are you going to show that someone got interrupted, or interrupted herself? Ellipses have a completely different feel.
The “find” command is so handy for this kind of super-boring nitpicky editing. But you know what I long for that I don’t have? I want a function that will show you every time you use the same word more than, say, twice in three paragraphs. I have no idea how you could write that kind of program, though, because obviously you’d need to exclude tons of little words: articles like “a” and “the”, pronouns, conjunctions. But it would be SO HANDY to be able to notice that you used the word “very” twice in two sentences, or had someone say something “quietly” three times on the same page, or all those other repetitions that look so horrifically stupid in the final version.
Copy editors do look for this kind of stuff, by the way. I just want something more foolproof than the human eye, even the skilled and dedicated human eye.
Anyway: done now! The WIP manuscript went off to my agent ten minutes ago. This the version that will go to my editor. Fingers crossed that she’ll love it!
8 thoughts on “The importance of tedious line-editing”
Tada! Here is a bookmarklet (tested in Firefox 18.0.2 only) which does an approximation of this. It isn’t difficult to add words to be excluded. As long as you can save your document as HTML and load it in a browser (even on your local machine), you should be able to use it.
Well, hey! Ask and you shall receive! Thanks!
And here is a bugfix version which hopefully doesn’t ignore words like actor, brigand, and anything containing ‘i’. Let me know if you spot any other problems, or want any tweaks to it.
This is embarassing, but here’s a second bugfix after I tried it on this blog post and noticed it was highlighting way too much. It maybe only affected the first couple paragraphs of a document, but obviously I’m too tired to reason properly about code right now. :)
Here is another version which should count the same word even if it’s capitalized differently (e.g. “sometimes” in your second paragraph).
Thanks for all these suggestions!
They are more like bugfix releases than suggestions. I suggest using the last one I posted. :) I saw your post, thought, “That sounds like fun! and it should only take about 10 or 20 minutes”, and here I am 20 hours and three revisions later. There is actually still at least one bug in the latest version (it won’t deal well with HTML markup inside your paragraphs because it just does a blind search-and-replace to highlight things), but the fix is several times longer than the program so far. (Although someone else already wrote it and released it without any restrictions, so I would have to do little more than copy it into my code.)
I don’t know if you use Microsoft Word or something else to write, but it would probably be possible to write a word processor macro that does the same thing, to avoid the ‘Save as Web…’ step. There are also analysis tools like Wordle that may be more or less practical or fun.
Yep, Microsoft Word. That’s entertaining — the fix being several times longer than the program! Isn’t that always the way!