Skip to content

Getting edited

David Weakliem and I wrote an article that will be appearing in American Scientist, which is a magazine, not a journal. Which means that the article is being edited by an actual editor (Morgan Ryan, in this case). It’s great: he read it over and made comments and changes, lots of things that make the article more readable and sensible.

It’s fun having an editor. I’ve written about 250 articles and 6 books, but this is the first time I’ve ever had a real editor on anything I’ve written.

P.S. And, no, a copy editor is not the same thing.


  1. Andrew Therriault says:

    As somebody who has spent time as both a copy editor and a social scientist (I worked in publishing for a few years before starting my PhD), I think you're a little too hard on copy editors.

    The job of a copy editor is a pretty crappy one, honestly. Every copy editor would like to be a "real" editor, as you describe them, making comments and changes to improve the overall quality of a book or article. Unfortunately, that's not how the publishing industry works.

    By the time a book gets to the copy editor, it's already been delivered to production, and the goal is to fix any glaring grammatical mistakes and turn around the chapters as quickly as possible (often in as little as 2-3 weeks for a 500-page book). While there's some allowance for queries and minor rephrasing, there's a lot of pressure not to either make significant changes (because authors can be highly defensive, and because real changes take time to circulate amongst the team) or give input on the overall quality of the text, even when such input could lead to significant improvement. And while your original text is probably quite good, most authors in scientific fields are terrible writers. It puts copy editors in a lousy position–either spend extra time (which is often uncompensated; budgets are set in advance) to get things right and be rewarded by incurring the wrath of overworked production editors who want to simply push a book through on a tight schedule, or else bite your tongue, fix the worst of it, and send it back and wait to get paid. Knowing that those impatient production editors are the ones who select copy editors for future jobs, which course do you think most copy editors choose, and can you blame them?

    This would be fine, of course, if acquiring or developmental editors had already gone through and done a thorough edit before it gets to the copy editor. That might have happened ten or twenty years ago, but not today. Most books in the sciences don't even get developmental editors, and acquisitions editors' job descriptions don't include line-by-line edits of their authors' work. Besides which, editors are typically evaluated according to the volume of the books they publish, not the quality of those books. So the incentive structure is totally perverse. It's ultimately a result of the increasingly-corporate structure of publishing; the roles are set by business people who've never had to actually make a book themselves, and the main focus is on short-term profits, not long-term quality.


    All that said–you may well have just had some bad copy editors. There's no formal qualification that makes someone a copy editor–the main requirement is simply that you know a production editor (or whoever else does the hiring) who likes you and thinks your English is okay. As often as not, copy editing jobs are done by production editors at one company for production editors at another, who then serve as copy editors on books the first production editor is making. It's not a bad deal if you can get it.

  2. Corey says:

    Minor tangent: in the article you've cited your paper with Francis Tuerlinckx for the concept of Type M errors, but that paper doesn't actually talk about Type M error at all. I know that I've read you on Type M error specifically, but I can't find the document now. What's the right cite?

  3. Andrew Gelman says:

    Andrew T.: Well, to be fair, I'm sure most readers of my books and articles are less concerned with readability than technical correctness–and an editor wouldn't help with the latter anyway. As it is, I feel that I write books to be read from beginning to end, but people probably just flip through them anyway.

    Corey: Type M errors are mentioned near the end of my paper with Tuerlinckx. I got the idea from Dave Krantz; apparently it's a well-known concept in psychometrics (but without the "type M" name; that was my idea).

  4. C. Zorn says:

    I'd agree, but add a caveat. Law review editors also act as "real" editors, but (in my experience) most of the changes they suggest (and often insist upon) actually wind up making the piece worse.

  5. Corey says:

    Whoops — "at all" was overstating the case. Still, I seem to recall reading an in-depth analysis of type M error akin to the one in the Tuerlinckx paper about type S error. Is my brain tricking me?

  6. Bob Carpenter says:

    I had a great copy editor for my first book from Cambridge University Press. She hugely improved the clarity of my sentences and even found mistakes in formulas at distances of many pages (to her, it just looked like I was being inconsistent — there were some false alarms here, too).

    There's a fine line between content and grammar. If the way you state a theorem is "technically correct" but ambiguously worded, it imposes a huge burden on the reader to resolve the uncertainty. Inconsistent terminology for the same concept is a typical case. Andrew's solution to the latter problem is to call every joint, marginal, or conditional density or distribution "p".

    Disambiguating is easy if you're an expert bringing top down knowledge to the problem. But for beginners, it can be devastating.

    As these uncertainties in a reader's mind pile up, they find themselves solving mental zebra puzzles ("Kools are smoked in the house next to the house where the horse is kept; The Norwegian lives next to the blue house; … Who owns the zebra?")

  7. Ken Williams says:

    I've written a book for O'Reilly & Associates, and their editing process was fantastic. They engage technical editors from the relevant field by contract, and their comments & suggestions made huge improvements for us.

Where can you find the best CBD products? CBD gummies made with vegan ingredients and CBD oils that are lab tested and 100% organic? Click here.