Thursday, August 04, 2016

Why are academics such bad writers?

Not all academics are bad writers but the exceptions are few and far between. Several recent articles in The Chronicle of Higher Education have attempted to explain why we can't write. There are two types of academic writing. The style you use in your academic papers differs from the style you use in writing for a general audience. There's absolutely no debate about the style of writing in the academic literature: it is horrible and it needs to change.

I want to talk about the other kind of writing; the kind where academics try to explain things to non-academics. I'll concentrate on science writing although I'm sure the same issues apply to history, philosophy, and all the other academic disciplines. I'm particularly sensitive to this problem since I'm working on a book about genomes and junk DNA and it's very different than writing a biochemistry textbook.

The latest (Aug. 1, 2016) article is an interview with Steven Pinker, the well-known Harvard psychologist. He's published seven trade books and is widely perceived to be a good example of how academics should write for a general audience [Scholars Talk Writing: Steven Pinker].

Pinker has already written about bad academic writing back in 2014 [Why Academics Stink at Writing]. In this recent interview he was asked to give the short version. Here's what he said ...
First, academic writers start off with the wrong tacit goal. Rather than trying to show their readers something interesting in the world (what Francis-Noël Thomas and Mark Turner call "classic" style), their main goal is to prove that they are not naïve about how terribly difficult it is to assert anything about anything in their field (what they call "self-conscious," "ironic," or "postmodern" style). In that defensive stance, they clutter their prose with hedges, apologies, shudder quotes, narcissistic observations about their profession (as opposed to its subject matter), and metadiscourse (discourse about discourse).

Second, academics suffer from the Curse of Knowledge — the difficulty of appreciating what it’s like for someone not to know something that you know. So they fail to explain their jargon, spell out their acronyms, or supply concrete details that would allow the reader to form visual images of what they’re describing.

Finally, they have little incentive to care. Good prose requires dedication to the craft of writing, and our profession simply doesn’t reward it. It isn’t taught in graduate school, and few reviewers will veto a manuscript or a grant application just because it’s a painful slog to read.
He's talking mostly about academic writing here. I agree with him about all three points. The last one only applies to papers published in the academic literature. The opposite criterion applies when writing trade books where good prose is valued far above good science.

The first problem is one I'm very familiar with. In trying to write a book about junk DNA, I'm terribly conscious about not deceiving my reader by dumbing down the subject matter. In fact, I've devoted many pages to explaining why it's so difficult to "assert anything about anything" in my field. You just can't simplify a complex topic without misleading your readers. That's why my writing is full of hedges, qualifications, exceptions, and distracting tangents and footnotes.

I follow one cardinal rule in writing these essays—no compromises. I will make language accessible by defining or eliminating jargon; I will not simplify concepts.

Stephen Jay Gould
The Flamingo’s Smile p. 16
It's so much easier to exaggerate and simplify by ignoring the messiness of biology. Stephen Jay Gould is one example of an academic who does not fall prey to this fault but the result is a level of writing that's way beyond the average reader and even beyond most scientists. This is one reason why he is criticized so vehemently: he writes about complex issues in a complex way so it requires a bit of effort to understand.

Richard Dawkins, on the other hand, is often admired for making complex issues seem very simple. Perhaps too simple.

The second problem (the "Curse of Knowledge") is related but different. Let me give you an example. I want to write about genes in order to make readers appreciate why 90% of our genome is junk. This means explaining what a gene is and describing two kinds of genes: protein-coding and those that produce noncoding, functional RNAs.

Why is this important? Because a lot of the debate over junk DNA depends on understanding whether transcribed regions of DNA are genes or not. That means readers have to understand transcription (and translation) and they have to understand what causes transcription of a gene; namely, RNA polymerase binding to promoters under the influence of transcription factors and regulatory sequences. You can't understand and appreciate the issues in the junk DNA debate unless you know about these things. Readers need to know why a stretch of DNA isn't a gene just because it's occasionally transcribed. That's a lot of "jargon" and details and I don't think my publisher will like it.

There's a middle ground somewhere that doesn't involve dumbing down to the point of misrepresenting the field but not including so much detail that you lose your audience. Finding that middle ground is much more difficult than most people realize. In my opinion, most science writers aren't even close to the middle ground—they sacrifice accuracy for style every time because that's what sells books.1

Steven Pinker has some advice for those of us who don't meet his standard. We tend to write too much. He says,
In part this bad habit comes from defensiveness: Writers fear all the possible objections and fend them off pre-emptively. In part it comes from self-presentation: the desire to flaunt one’s erudition and justify one’s history of reading and research. And in part it comes from incompetence.
The problem is real but I think Pinker is wrong about the motive. In my case, I want to fend off objections and present the other side of argument because I value critical thinking and honesty; not because I want to flaunt my erudation or justify my scholarship. I'm trying to do the same thing I expect of my students.

Pinker continues,
A well-structured essay carries the reader along without a lot of signposting. But if the essay is structured in the order in which thoughts occur to the writer, he or she will have to erect obtrusive previews, summaries, and signposts to prevent the reader from getting lost. A lack of attention to concision can fatten prose at every level of organization. Strunk and White’s prime directive is to "omit needless words" (a lovely example of itself). But it takes skill and effort to spot and extirpate the needless words — and the needless sentences, paragraphs, and sections.
We all agree that it takes skill and effort to write concisely without sacrificing accuracy.

I'm not a big fan of Steven Pinker,2 even though he's a Canadian, but I know he's right about how difficult it is for academics to write for a general audience. Let's see how he puts his recommendations into practice.

I direct your attention to another essay in The Chronicle of Higher Education from just a few weeks ago (July 11, 2016). This one is on the same topic. The author is Noah Berlatsky, a non-academic: Why Most Academics Will Always Be Bad Writers.

Berlatsky explains that academics will always be bad writers because they are academics. Here's what he says about academics ....
Writing is a skill, and — as any editor will tell you — it’s not one that everyone possesses. Academics are primarily researchers and teachers; there’s no reason those talents should necessarily overlap with writing. To my mind, the real surprise isn’t that so much academic writing is bad, but that so much of it is comparatively well written and entertaining.
I find this comforting! It means that my lack of skill doesn't necessarily interfere with being a good academic!

Berlatsky, like me, isn't a big Pinker fan. He illustrates the problem of over-simplifying by using an example from one of Pinker's own books.
Steven Pinker himself has on occasion simplified his message in unfortunate ways. In his hugely successful 2011 volume Better Angels of Our Nature, for example, Pinker puts forward the thesis that humankind has become less and less violent. To support this argument, he writes: "The worst atrocity of all time was the An Lushan Revolt and Civil War, an eight-year rebellion during China’s Tang Dynasty that, according to censuses, resulted in the loss of two-thirds of the empire’s population, a sixth of the world’s population at the time."

That is a perfectly clear and precise sentence. It’s also misleading to the point of being an outright falsehood. As Pinker says, he’s extrapolating from census data. But you can’t treat 8th-century censuses as some sort of straightforward registry of wartime death tolls. One expert on the population statistics of China notes, "Even if such a huge loss were conceivable, it would be naïve to suppose that an accurate count could be carried out in the midst of the ensuing chaos." Other researchers have tentatively placed the death toll at something more like 13 million — though even that’s very dicey. The truth is we don’t know for sure how many people were killed in the An Lushan rebellion. To be accurate, Pinker would have had to have been vague. The rage for clarity led him astray.

I’m not trying to impugn Pinker: Anyone can make a mistake, especially when writing a book like Better Angels of Our Nature, which attempts to synthesize a vast amount of information from a wide variety of fields. But that’s exactly the point. It’s not easy to communicate complicated data and ideas with precision, style, and a modicum of propulsive punch. Many professional writers stumble into infelicities and inaccuracies. Why should academics be any different?
Think about this problem the next time you listen to a TED talk or read a book or a article in the poplar press written by an academic. If it's easy and pleasant to read and the problem seems so simple that everyone can understand it, then, chances are, it's not accurate. That's not how academics write.

I'll close by mentioning how Richard Dawkins handled this problem. He wrote a book called The Selfish Gene—a book where he advocates a new way of looking at genes and evolution. There was criticism, so Dawkins wrote a second, more academic, book in 1982. It sold far fewer copies. I suspect most of you haven't read it.

Here's what Dawkins wrote in The Extended Phenotype (p. 8).
Parts of some early chapters are frankly retrospective and even defensive. Reaction to a previous work (Dawkins 1976a) [The Selfish Gene] suggests that this book is likely to raise needless fears that it promulgates two unpopular '-isms'—'genetic determinism' and 'adaptationism.' I myself admit to being irritated by a book that provokes me into muttering 'Yes but ...' on every page, when the author could easily have forestalled my worry by a little considerate explanation early on. Chapters 2 and 3 try to remove at least two major sources of 'yesbuttery' at the outset.
I suppose that's one way of handling it. Write the first book for style and flair then publish a second one to deal with all the complications and yesbuttery you ignored in the first book!

I'm going to try and avoid yesbuttery in my book. I do so in the full knowledge that it will make the book less readable, and less accessible, to a general audience. Hopefully, it will be more accurate. It won't be good writing because I'm an academic!

1. I fear that the problem with many science writers is not that they are sacrificing accuracy for style—it's that they don't know what's accurate in the first place.

2. See: Steven Pinker defends "neo-Darwinism," whatever that is and Changing Your Mind: Are Humans Evolving?.


  1. Stephen Jay Gould is one example of an academic who does not fall prey to this fault but the result is a level of writing that's way beyond the average reader and even beyond most scientists. This is one reason why he is criticized so vehemently: he writes about complex issues in a complex way so it requires a bit of effort to understand.

    I liked a lot of what Gould wrote, but if anything I think the opposite criticism of him was more common -- that his description of biology was too simple and that he glossed over the complex mathematical aspects of evolution such as population genetics (perhaps because he wasn't that familiar with them). He did like to use what my grandfather used to call "$10 words" though.

    1. I have heard that he had a deal with Natural History Magazine that he would not be subject to any editing whatsoever. That sounds great, but what it led to was that he became more and more wordy. A review of his late book The Structure of Evolutionary Theory was rude about his prose and said "can you say logorrhea?"

    2. Yes, his later essays run on and on. I liked the earlier ones, though, a lot.

  2. I'm not a fan of stephen Pinker either but the point about hedging, humming i find very real in the great number of pappers i have read on geomorphology.
    Its like they are afraid to commit even though they are writing to make/offer new ideas with punch. Yes afraid.
    i think writing is a skill also and agree that considering they are not writers its not that bad. Science types, on a curve, are sharp enough to write well though not writers.
    It is a skill. people complain about my 'skill' even though i write as I think and speak. 9if that helps explain it).

  3. I think part of the problem is trying to make an idea seem important, by which I mean, something that makes a difference in everyday life. Or something that changes the way we look at the world. Bestselling writers do this, possibly at the expense of exaggerating.

    Titles reflect the desire to capture the imagination, and titles often scream.

  4. I once got a grant review that said "the proposal is clearly written" followed by comments that indicated the reviewer regarded that as a negative.

    1. A friend was asked to show how she would present a topic to beginners. She made a very simple, clear presentation. One of the people on the committee thought she shouldn't get the job because she didn't know the materal well enough.

      Fortunately, the other committee members knew her (from before the interview), understood what she'd done and why, and insisted on hiring her.

  5. Recently, we were told that we must refer to ourselves in the plural when writing a single author paper, so we did. But now look what's happened to us.

    Almost all the biology we know is self-taught. For molecular biology, that mainly means Alberts et al, which we think is a great text book. Yes, it's a 1300-page monster, but there are more pictures than pages. It's odd, and sad, that most books aimed at the general public have so few.

    We like the notion of Lies-to-children ( "We happily teach children that 'you cannot take 3 from 2' because we are confident that someone will later introduce them to negative numbers." If you're writing for adults you can't be confident they'll ever revisit the subject you're writing about, so you have to be much more upfront about your lying, but we still think it's a good strategy.

  6. I had to write a business case while back; my grad student and I were looking for a grant to develop an crazy idea. The grant was aimed at commercialisation, and not science, although the money would be spent on R&D. I hated every minute of the process.

    The reviews from the innovation office at the university kept coming back to dumb it down. But there came a point when it was just full of half-baked analyses, market research pretty much made up whole cloth, and pie-in-the-sky promises.

    The dichotomy of the two styles of business writing and science writing really hit hard. It really felt awful to write such dreck and a finish a series of related research papers at the same time.

    Under the guidance of the innovation office, it got funded. Take a step back and the final result was pretty slick, though the old adage about sausage making really applied.

  7. Comprehensible and comprehensive.

    Pick one.

  8. Not all the academic writing services suck, just most of them are typical and its seems like copy paste