Steve Pinker on how to write science

November 16, 2012 • 9:10 am

I have it on reliable authority that Steve Pinker’s next book will be on modern grammar and usage: a Pinkerian update of Strunk and White’s famous The Elements of Style (a book I wore out with frequent use, but learn from the lecture below is flawed). And Steve’s already giving talks about this book to come.

I would have thought that after finishing his 832-page monster, the superb book The Better Angels of our Nature, Steve would have taken a breather, but if you know him you’ll realize that’s not on—and it makes me envious! He’s a book-writing machine, but all his books are engaging and well written.

At any rate, Steve has previewed his book in a new talk, delivered at the Massachusetts Institute of Technology (MIT), called “A sense of style.”  You can see the 77 minute video here.

As the MIT site describes, this is “The first annual Department of Nuclear Science and Engineering lecture about communicating complex scientific and technological subjects clearly and engagingly in the series: ‘Communicating Science and Technology in the 21st Century’.”

And it’s a very good talk.  If you have students, show it to them; if you write yourself, watch it. It’s not just about communicating science—to both the public and our colleagues—but about how to write clearly on any nonfiction topic.

His emphasis is on the “classic style,” a style I’ve tried to achieve (without copying its adherents), and of which I first became aware (do not end sentences with prepositions!) by reading Richard Dawkins. Note how Steve compares Dawkins favorably with the world’s most opaque and infuriating academic writer, the postmodernist Judith Butler (see Martha Nussbaum’s magnificant takedown of Butler in The New Republic).

In that style, one paints tangible pictures for the reader but never condescends, and writes as if the text should be read aloud (the “conversational style”); those are valuable tips. Steve’s examples of good and bad prose are enlightening, and his delivery instantiates his own clear but personal style. But watch the video yourself; you won’t regret it.

A screenshot of one of his slides:

h/t: Chris

72 thoughts on “Steve Pinker on how to write science

  1. One of the main reasons I take the time to read your articles on this website is that I enjoy your style of writing. No need to be too envious of others.

  2. By way of subscribing, let me state that I don’t think you have to worry about your own writing style, Jerry. Of course, keep polishing it; all skilled artisans never stop pushing themselves closer to perfection. But you’re one of the exemplars.

    One of your here-named idols even confirmed as much on the jacket of your book….


    1. This cannot be over-emphasized: If you care about your craft, you never “arrive.” There is only better (and sometimes worse …).

    1. I think the many, repeated criticisms of Strunk and White are largely overblown. I have been teaching a senior-level evolution course for many years. It has a significant writing component, and we are meant to judge student review papers as much on the basis of writing quality as on the basis of scientific rigor, timeliness, etc.

      Given the rather low STARTING POINT of MANY of my students in terms of writing ability, S & W is an extremely useful, concise guide. I agree with some of the criticisms of this “little book,” but it STILL does the job for students who simply have a long way to go. Moreover, the book appears to have served Jerry Coyne well. If it were really as odious as some claim, I suspect that Jerry’s writing would have actually deteriorated as he wore out his copy through frequent use.

      1. Did you read the article I linked to see its specific criticisms? Also, did you stop to think that despite Dr. Coyne’s anecdote that his good writing is a result of his mimicry of his reading materials, whether conscious or unconscious, as well as his extensive experience writing?

      2. As soon as I saw that Jerry mentioned Strunk and White, my first thought was where’s Geoff Pullum? I’m so glad that jaxkayaker linked to his essay. Talk about this little book is a sort of Shibboleth, and I can’t wait to hear what Pinker, particularly with his knowledge of linguistics, has to say.

        I, too, have the reaction of wondering if you actually read the short piece written by Pullum. For sure, he’s been dealing with this issue almost as long as Eskimo words for snow, but I think he clearly and succinctly advances his cause. It’s a little David and Goliath, but the facts are on his side.

        Strunk and White isn’t a horrible style guide. When it comes to writing advice, there are probably worse books out there. The point, though, is at what cost? Geoff Pullum’s answer is that if you actually read the book and have any knowledge of how the English language works your head should probably explode. Style suggestions are not the same as explaining why something works. And yet that’s exactly what Strunk and White do, incorrectly, often enough. We’re talking about two guys who can’t properly identify a passive clause telling your students to (sometimes) avoid writing them. How useful it that?

        There are better style guides out there, where either the authors don’t pretend to understand the nuances of English linguistics or they simply accurately know how and why things work. Should we keep using this book if it does help science students write better? I think not, since using Strunk and White invariably puts your students into contact with incorrect explanations of and false information about English grammar.

    2. Agreed on it being a myth, but I still wince when my sentences happen to end with one. Arguably, there are stronger ways to end a sentence.

    3. IIRC, Geoffrey Pullum (who, like almost all academic linguists, is a hardcore descriptivist) has railed against not just Strunk & White, but William Zinser, Orwell’s essay “Politics and the English Language,” and anything else that has even the faint whiff of prescriptivism upon it. (FWIW, I think Pullum is right about all of this and find his writing to be incisive, eloquent, and at times acerbically funny. I have derived many hours of pleasure wondering among his articles and those written by his fellow contributors at their Language Log website.)

      But at least while on that Language Log site, Pullum occasionally adopts the air of a guy who walks into a bar looking for someone to hold his jacket so he can start throwing punches (or, in the academic blogosphere equivalent, so he can threaten to refund of the Language Log subscription fee of anyone who has chanced to offer a comment that raises the Pullumian ire). At such times, he can be overly harsh on anyone venturing advice to would-be writers trying to hack their way through the thickets of English composition.

      Inasmuch as I’ve never paid a Language Log subscription fee, I feel (perhaps foolishly) immune from any retaliation for making this criticism.

      1. That’s “wandering” about — Jesus, don’t want to make a usage mistake before climbing back off this limb.

  3. The idea that you should not end a sentence with a preposition is a not a “rule” of composition. It’s a matter of style. Among other arbitrary claims (like the assertion that you should not split an infinitive), it was first proposed in 1762 by the clergyman and amateur grammarian, Robert Lowth, in his “A Short Introduction to English Grammar,” a book whose unwarranted influences have led to all kinds of needless clumsiness in student writing. It’s a myth, yet owing to the indoctrinating attitudes of too many poorly informed grade school teachers, it endures. There are many occasions in composition when placing the preposition at the end of the sentence will help the writer avoid wordiness, awkwardness, and archness in tone, for example, “Do you know where that came from?”

    In MODERN AMERICAN USAGE (Oxford University), usage expert Bryan A. Garner calls this one a “superstition”: “The spurious rule about not ending sentences with prepositions is a remnant of Latin grammar, in which a preposition was the one word that a writer could not end a sentence with. But Latin grammar should never straitjacket English grammar. If the superstition is a ‘rule’ at all, it is a rule of rhetoric and not of grammar, the idea being to end sentences with strong words that drive a point home. That principle is sound, of course, but not to the extent of meriting lockstep adherence or flouting established idiom.”

    1. The idea that you should not end a sentence with a preposition is a not a “rule” of composition.

      “This is a rule up with which I will not put.” W Churchill (apocryphal).

      1. Steven Pinker to boldly go where Strunk and White were loath to tread.

        “There is a busybody on your staff who devotes a lot of time to chasing split infinitives: I call for the immediate dismissal of this pedant. It is of no consequence whether he decides to go quickly or to quickly go or quickly to go. The important thing is that he should go at once.”

        (George Bernard Shaw, Letter to The Times)

      2. In addition to the apocryphal Churchill quote, the other locus classicus of the rule against the rule against ending sentences with prepositions seems to be the story about the Ivy League professor who bellies up to a red-neck bar in Texas. The guy in the next seat pushes back his ten-gallon hat, looks him up and down from the elbow patches on his tweed jacket to his weejuns, and asks: “Where you from?”

        “I, sir, am from a place where we know better than to end a sentence with a preposition.”

        “So, where you from, asshole?”

    2. Perhaps you’ve heard the story about the little boy who sees his father coming upstairs with the wrong bedtime storybook and says “Dad! What are you bringing that book I don’t want to be read to from out of up for?”

      (Paraphrased from David Carkeet’s novel The Full Catastrophe.)

    3. It’s also superseded by one of the other ‘rules’: eliminate unneeded words. The reason “Where are you at?” is less grammatical than “Where are you?” is not because the former ends in a preposition, but because the preposition is unnecessary.

      1. I should add that “Where are you from?” is not less grammatical than “Where are you?” because word ‘from’ gives the whole phrase a different a meaning.

    4. Man in bar, to his neighbour: Where’re you from?
      Neighbour: Somewhere where we don’t end sentences with prepositions.
      First man: Sorry. Where’re you from, asshole?

        1. Ken Kukec beat me to it though. That’s the trouble with ordering comments by who’s answering who, it’s easy to miss things. (Or ‘answering whom’, for the pedants.)

    1. I think it’s US usage that has created many unnecessary add-ons, such as orientates for orients, obligates for obliges, elevator for lift, and the many tautological redundancies (or redundant tautologies) that clutter US academic writing. (Frederick Crews sent those up in The Pooh Perplex with “refutatory elenchus … that elegant expression.”)

      1. Whilst I agree in general, I think you’ll find that both “orient” and “orientate” are considered correct, with the former being more popular in the US, and the latter in the UK.

        “Tautological redundancy” is tautological! Sorry if that was intended as a joke.

  4. As an engineer I thank you. In my experience it is hard to remember such advice when you’re mired in the struggle to write a memo, but I think one idea that may stick in my mind is Steve’s admonition to point the reader to something concrete they can “see” and then go from there.

  5. Pinker is one of my favorites, although I know some criticize him for (among other things) being a “popular” science writer in a pejorative sense. I appreciate his style, because he makes diffifult concepts clear — and I don’t feel talked down to at all. Quite the opposite — his prose, references, and anecdotes delight the mind. I found the same true of Coyne’s book, and the one he recommended by Donald Prothero. They make their specialized training accessible to people like me whose academic backgrounds lie elsewhere.

  6. I still like The Elements of Style and it can help a lot of writers (well, a good editor would help them even more.)

    I also found very useful: On Writing Well by William Zinsser.

    Mainly, you must read a lot to write well. And be willing to edit yourself or subject yourself to another editor.

    1. Exactly. Most publishing writers work closely with editors. Skilled and facile as he is all on his own, I’m sure Stephen Pinker has worked with an excellent editor at Viking, just as Jerry must have done at the same house for WEIT.

    2. “And be willing to edit yourself…”

      What was Dr. Johnson’s advice? “Read over your compositions, and where ever you meet with a passage which you think is particularly fine, strike it out.”

    3. I don’t have my copy where I currently live, but if I remember correctly, Zinsser’s book was handy, and less objectionable than Strunk and White. The problem with their work, which I mentioned above, is that their discussion of style is cloaked in the guise of a grammar lesson. That’s not a problem, necessarily, except that in their case the information about grammar is often enough inaccurate.

      As far as I can tell, reading and thinking are, as you note, particularly effective in leading to good writing.

  7. There’s a lot I appreciate about this post — not just the response to Pinker but also the Butler dig. However, I’d still like to see Pinker work on his slide design.

    1. The dig at Judith Butler’s obscurity –

      “The move from a structuralist account in which capital is understood to structure social relations in relatively homologous ways to a view of hegemony in which power relations are subject to repetition, convergence, and rearticulation brought the question of temporality into the thinking of structure, and marked a shift from a form of Althusserian theory that takes structural totalities as theoretical objects to one in which the insights into the contingent possibility of structure inaugurate a renewed conception of hegemony as bound up with the contingent sites and strategies of the rearticulation of power.”

      – is a classic, and I was happy to be reminded of the late Denis Dutton, another very clear writer and thinker, who brought it to light – if that is the expression.

      Presumably there are post-modern thinkers – the people Butler was writing for – who can read the quote and immediately understand it. I wonder if it has ever been translated into plain English?

      Pinker’s discussion of the Theory of Mind is interesting, both in its application to writing to add to a reader’s knowledge, based on what can be assumed about what they already know, but also in ordinary life. As a borderline Asperger, I have always suffered from my assumption that others know what I do. It was reassuring to know that neurotypicals have this problem too.

      1. “Presumably there are post-modern thinkers – the people Butler was writing for – who can read the quote and immediately understand it.”

        Just as presumably there are imperial courtiers who can immediately see, and appreciate the exquisite weave of, the Emporor’s New Clothes.

        1. So are you saying that those who claim to understand it are lying, or just self-deluded? This is a testable claim. At least I thought it was until I started trying to work out how to test it. If one person translated it into Russian, another translated it back, and a third judged that the meaning was substantially the same, would that verify its coherence? Help me, I’m struggling here….

          1. I’m saying that, while the words are cognizable, they are nearly content free. They’re being used not so much to convey ideas, as to demonstrate in-group membership among the cognoscenti.

            Although there have been worthwhile ideas promulgated by the post-structuralists, especially early on, as it spread as a fad through academia, it picked up adherents more interested in striking a pose, and its founders and leaders came to regard themselves as legitimate geniuses, keepers of the flame, whose every utterance merited not just detailed study, but adoration — such as the court physicians bestowed on the bowel movements of the child emperor of China. (This loop back to imperial simile is coincidental.)

            As to whether its practitioner and adherents are lying or self-deluded, it’s my policy never to accuse anyone of lying without irrefutable proof. Never attribute to malice that which can be explained by incompetence, negligence, or mistake. I suspect some practitioners are churning out prose knowing it is essentially content free, and that others have deluded themselves into thinking momentous ideas are being bandied about. But as stated above, the primary purpose of this writing is not to convey ideas, but to indicate status. (Ironically, the use of language for such purposes is among the early, worthwhile insights the deconstructionists illuminated.)

            Regarding your proposal to test it for coherence by translating passages into a foreign language and back, it’s an interesting idea that might yield some insight. But I’m not sure it would prove what you’re looking for, since it might be possible to accurately decode the words into each language, such that they’re equally incoherent in both.

            In any event, much of post-structuralists writing already sounds as though it’s been transliterated by machine. (To be fair, much of what I’ve read was already in translation from the French.) But even when written originally in English, as in the Judith Butler passage quoted above, the writer seems more interested in flaunting a facility with jargon, than in communicating ideas to a reader. The words in her sentences clang tonelessly against one another, like boxcars banging about in a railroad switching yard.

          2. In response to your request regarding a test, I would propose the following: These writers presumably began as persons fluent in plain English (or plain French, or plain whatever) before winding up as high-wizards of academia. They ought to be able to recapitulate their journey for the interested layman, to communicate the basic ideas of their field (putting aside whatever esoteric nuance can only be communicated through the jargon of the initiates), just as they had to acquaint themselves with these ideas, and then initially formulate their own concepts, using a more-basic version of English they then had available.

            Our science popularizers, the best of them, are able to convey concepts as arcane and complex as quantum mechanics, special and general relativity, and higher mathematics using straightforward prose accessible to reasonably intelligent and motivated laymen. I see no reason to expect less of those in the “humanities” (if we may deign to include what the post-structuralists/deconstructionists/post-modern philosophers of language do within that rubric). So if there are worthwhile ideas buried in the cant of post-structuralism, why can’t its practitioners explain those essentials to us in language stripped of its over-stuffed baggage?

      2. Probably the same people who bought “Transgressing the Boundaries: Towards a Transformative Hermeneutics of Quantum Gravity.”

    1. Me too, though I think of it as Fowler. (Modern English Usage by H. W. Fowler, 1926, revised by Sir Ernest Gowers, 1963) I could read it for pleasure. Much in it is now antiquated, but t/he/y always gave reasons for his/their advice that were good for their time.

      I look forward to Pinker’s book when it comes out.

  8. I’d love to watch this video, but my ‘rural’ connection is not up to it – watching 77 minutes of a talk in 15 second grabs is far from ideal.

    Anyone know of a download link or alternate hosting on a site which does better pre-caching?

  9. I think we need more Fowler not less, not because the language is, or should be, static, but because when young editors at daily newspapers can’t tell the difference between its and its, or, risibly, coiffed and quaffed,, there’a crisis of meaning brewing.

    As for Butler, her sort of academic cant may mean something, but it disguises it so well. I usually stop reading at the first appearance of :hegemony” or “hermeneutics.”

  10. Felt called out around 30:15. I think. I guess. I suppose. It seems to appear in the majority (under some circumstances) of instances with exceptions. I am definitely guilty of this hedging, there’s an intuition on my part that this more accurately reflects the truth. It never occurred to me that it might also be annoying.

    My Aha moment was around 31:29 though: “…rather, you write as if they are true, and you count on the reader to fill in all of the missing hedges and apologies and qualifications and so on.”

    I think Dr. Pinker is putting an admirable amount of trust in the reader, but perhaps (hedge) that memo has not been distributed on a widespread basis. I don’t disagree that this is how things should work, but I think that when you see vitriol start to boil in a debate it often comes down to a presumption that “So and so is acting as if they absolutely know X! How dare they!”

    I think training on scientific writing is a beautiful thing, but training on scientific reading might be more desperately needed (self included) in society as a whole.

  11. I was struck by this sentence in a paper recently published in Molecular Ecology:

    “To examine the effects of the degree of habitat connectivity of patterns of population differentiation, projections from current species-specific SDMs were used to define populations.”

    (where SDMs are species distribution models, i.e. ecological niche models)

    That’s about six nested levels of abstraction/reification before the main clause, and several more in the subject noun phrase before we get to the passive verb.

Leave a Reply