"Liven up your results by reporting them in furlongs, chaldrons, and fluid scruples."
March 24, 2012 1:03 PM   Subscribe

 
A major contributor to the dull language of research is the fact that papers need to be readable by an audience of non-native readers. Technical jargon can be learned. "Dorsolateral prefrontal gyrus" is an unambiguous entity. One could argue that the language that couches the jargon should also be unambiguous, unpoetic, and unvarying. If "only" works in a given context, why belabor readers with synonyms like "lone"?
posted by Nomyte at 1:13 PM on March 24, 2012 [12 favorites]


(Or, to rephrase that last sentence in the passive voice, as seems to be the scientific fashion, why must dryness be written by us?)

On the whole, I think good ('non-scientific') prose is distracting in scientific papers. But I am so, so, so with this guy on the passive voice thing.
posted by gurple at 1:14 PM on March 24, 2012 [1 favorite]


Why can’t we write like other people write? Why can’t we tell our science in interesting, dynamic stories? Why must we write dryly? (Or, to rephrase that last sentence in the passive voice, as seems to be the scientific fashion, why must dryness be written by us?)

Because, sadly, most scientists (and even most academics) are not very good writers, and so any attempt to write well will stand out (and possibly make other people look bad) so it must be stamped out.

On the other hand, I find the prose of Robert Folk, geoscientist extraordinary, to be fairly amusing (in a dry sort of way), so there are exceptions. If only there were more, however.
posted by GenjiandProust at 1:15 PM on March 24, 2012


And, because I cannot resist:

Metafilter: For some reason, Vladimir Putin.
posted by GenjiandProust at 1:17 PM on March 24, 2012 [3 favorites]


I sympathize with the author so goddamn much. I've started paper number two and have an additional four I'm supposed to finish by the end of next year. Every time I revise a draft I weep as I prune out any sense of enjoyment in the English language. Obviously, you shouldn't wax poetical for a paragraph if you can write a concise version of the same thing, but you don't have to be flowery by any means to maintain some variety. A word like 'lone' does absolutely no harm when it's replacing one of the fifteen 'only's.

Also #12 is so, so accurate.
posted by six-or-six-thirty at 1:18 PM on March 24, 2012 [1 favorite]


As far as I'm concerned, if it was hard to write it should also be hard to read.

(I joke. Maybe. Want to see my dissertation?)
posted by LastOfHisKind at 1:20 PM on March 24, 2012 [3 favorites]


Anybody else hearing Guided By Voices?
posted by jonmc at 1:23 PM on March 24, 2012 [3 favorites]


Worth reading for #12 alone. Six-or-six-thirty speaks the truth - this is pretty much exactly how it is done.
posted by Mitrovarr at 1:25 PM on March 24, 2012


How to rhyme like a scientist
posted by fuq at 1:28 PM on March 24, 2012


Try and use Furlongs as a unit in one of your papers, you say? Challenge accepted!
posted by redbeard at 1:38 PM on March 24, 2012 [1 favorite]


I want to see fph used (furlongs per hogshead) in the same paper as feet per hour.
posted by BrotherCaine at 1:50 PM on March 24, 2012


Science succeeds in spite of human beings, not because of us, so you want to make it look like your results magically discovered themselves.

I sympathize with this notion that perhaps there would be clear benefits to humanizing scientific output. We should remain keenly aware that scientists are mortals, too, subject to the same vagaries as everyone else. Whether it's a pet agenda, a grudge, a stubborn defense of reputation, people don't cease being people once they go through a research ethics course. I personally think that this "Decline Effect" is the result of cutthroat publish-or-perish academic job culture. At a certain point, the pursuit of reputation supersedes the pursuit of knowledge. Sigh.
posted by stroke_count at 1:54 PM on March 24, 2012 [1 favorite]


Effective communication is a tough nut to crack, especially in science. We need more hand gestures in our journals.
posted by Blazecock Pileon at 1:58 PM on March 24, 2012


How do you get into Nature?
posted by bukvich at 2:04 PM on March 24, 2012


"But in his mind, “lone” must have conjured images of PvPlm perched on a cliff’s edge, staring into the empty chasm, weeping gently for its aspartic protease companions."

This is the funniest sentence I've read in a long time.
posted by anaximander at 2:08 PM on March 24, 2012 [12 favorites]


Every time I revise a draft I weep as I prune out any sense of enjoyment in the English language.

Oh my god yes. I'm a programmer in a lab, so writing papers is not technically in my job description, but it's a responsibility I accepted in the mistaken belief that it would be a pleasant diversion from my usual work. Now I would kick a million puppies to avoid any further writing assignments.
posted by invitapriore at 2:23 PM on March 24, 2012 [2 favorites]




I once wrote a paper and handed it to my PI at the time. He looked at it and said that it wasn't professional enough. He suggested that I read the writing of a guy in a related field, saying that this guy could write pages and pages of discussion without saying anything. I internally wept. Then I went and got pdfs of this guy's papers, and listed all the "science" words and phrases. Then I liberally sprinkled them into my paper. I used the phrase "shed light upon", fer chrissakes. The resulting mess was returned to my PI. He loved it. We published it with limited revisions.

I've still got that list somewhere if anyone wants it for sprinkling purposes.
posted by sciencegeek at 2:35 PM on March 24, 2012 [5 favorites]


Has anyone read old biology/ecology papers? They are full of this sort of English. They're amusing to read so some times I do it for fun.

But have you tried to pluck any useful information out of these papers? Horrible.

Science writing may be dry and impersonal but it is precise and can be replicated. I would hate to have science turn into a competition for who can write the best.
posted by hydrobatidae at 2:45 PM on March 24, 2012 [2 favorites]


This is one of the many reasons I got the best math and science education the Illinois tax payers' money could buy in high school but still ended up a filthy humanities undergrad.

But I still understand why scientific papers must be written the way they are, and those who can do it well have my deepest respect.
posted by MCMikeNamara at 2:53 PM on March 24, 2012 [1 favorite]


Starting sentences with “obviously” or “as everyone knows” demonstrates your intellectual superiority.

I also remember the word "clearly" signaling that the next thing the author is about to discuss will not be explained at all, leaving a huge knowledge gap for a reader new to the subject. That was especially fun in college.

Whenever I read the word "obviously," it's always in the voice of Severus Snape.
posted by hoppytoad at 3:05 PM on March 24, 2012


These results clearly demonstrate
posted by sciencegeek at 3:08 PM on March 24, 2012 [2 favorites]


I've read some engaging scientific (and mathematical) papers that conform to the norms of scientific writing. It can be done. My concern is that so many papers are poorly written, and it's difficult to glean information from a poorly written paper no matter the style of writing. Sometimes it makes me want to take the paper-writer by the shoulders and shake them while saying loudly and sternly, "do you not realize that the purpose of writing papers is communication?!"

I think a large part of the problem is that the incentives are all for churning out large quantities of papers, not for taking the time to edit and write well-written papers.
posted by eviemath at 3:11 PM on March 24, 2012 [1 favorite]


Clearly this is the appropriate thing to read while I procrastinate on my article review matrix.
posted by cobaltnine at 3:12 PM on March 24, 2012


This guy who was suppose to teach me how to write scientific papers made a big deal of using active voice and his idea of how to write a paper.
All that really matters is clarity.
Write for your audience, and that doesn't have to assume they are dry or ignorant, but they may be.
posted by provoliminal at 3:14 PM on March 24, 2012


I sympathize with the author so goddamn much. I've started paper number two and have an additional four I'm supposed to finish by the end of next year. Every time I revise a draft I weep as I prune out any sense of enjoyment in the English language. Obviously, you shouldn't wax poetical for a paragraph if you can write a concise version of the same thing, but you don't have to be flowery by any means to maintain some variety. A word like 'lone' does absolutely no harm when it's replacing one of the fifteen 'only's.
Well, as others mentioned, you may send a lot of foreign born grad students to their dictionaries. Who knows.

It would be nice if he'd actually tried to find out what the actual argument against words like 'lone' might be. Might they make the papers harder to understand in some cases? Would they take longer to read?
posted by delmoi at 3:15 PM on March 24, 2012


I think what the author is trying to get at, but not articulating, is that technical science writing is devoid of human spirit and literary considerations, and this comes at a social cost. I agree because art and science are really two sides of the same coin, and many of the best (classic) papers in various fields do demonstrate that it is possible.

The problem we have is that most researchers treat formal technical communication as a chore rather than an integral part of their craft. This isn't just because writing is hard, but because the external incentives to write favor a kind of short-sighted cold efficiency. This mode of communication is a direct result of compartmentalization and excessive competition that is so prevalent in professional research. What scientists face is a process of trying to make a living inside system where they are obliged to publish regularly, spend way too much time seeking grants, and overall navigating a social hierarchy. This is the age-old difference between survival and self-actualization, and the deliverables reveal this.

So yeah, while the article is correct, but I think this is the bigger picture to it.
posted by polymodus at 3:16 PM on March 24, 2012 [3 favorites]


The other thing to consider: many scientists may be bad writers. By avoiding all flourishes, you're giving them less rope to hang themselves.
posted by delmoi at 3:17 PM on March 24, 2012 [3 favorites]


It would be nice if he'd actually tried to find out what the actual argument against words like 'lone' might be.

Yup. The "problem" with "lone" is personification of the object of study. So it's not so much that it is unscientific diction, but more that science generally has a problem with personifying things. (whether or not that is valid).
posted by polymodus at 3:19 PM on March 24, 2012 [1 favorite]


One of my professors would dislike 'lone' for its connection to 'lonely.' A plasmepsin can't be lonely, as it doesn't have human feelings and emotions, and to use 'lone' to describe it is to athropomorphize. Just like you can't say that the oil doesn't want to touch the water because oil can't want.
posted by sarae at 3:20 PM on March 24, 2012 [2 favorites]


There are scads of flourishes in scientific writing. They are just the approved flourishes. You can't say that you're showing something, you have to say that you're demonstrating something. An experiment was not done it was performed. Clearly we see here that the focus of Bob's study fails to provide the crucial evidence needed to demonstrate what our findings capably demonstrate.
posted by sciencegeek at 3:21 PM on March 24, 2012 [2 favorites]


Sole works, right.
posted by sciencegeek at 3:21 PM on March 24, 2012 [1 favorite]


One of the biggest reasons why this is the case is that too many scientists couldn't write in such a way as to be easily read if they wanted to. They also, for the most part, don't know it and the only thing worse than dry language is a swampy wet mess. By imposing cultural restrains on the writing of our peers we are allowing the more writing challenged among us to be heard. Also as mentioned up thread, most of the people who need to read the English poetry this guy really wants to write don't speak English as a native language.
posted by Blasdelb at 3:23 PM on March 24, 2012


It's a good thing mathematicians aren't scientists, or I'd take issue with this article's thesis. This person has never read Fruit Salad [pdf] by Gyarfas or The Two Cultures of Mathematics [pdf] by Gowers.
posted by King Bee at 3:27 PM on March 24, 2012 [1 favorite]


sciencegeek: "There are scads of flourishes in scientific writing. They are just the approved flourishes. You can't say that you're showing something, you have to say that you're demonstrating something. An experiment was not done it was performed. Clearly we see here that the focus of Bob's study fails to provide the crucial evidence needed to demonstrate what our findings capably demonstrate."

I apologize if I seem Vogonesque, but as far as I can tell there is an approved flourish for each necessary purpose. Having a smaller number of flourishes makes the writing dull and a little painful but it allows for clarity from shitty writers, reduces the barriers to entry for both reading and writing papers for the vast majority of scientific workers, and it gives us a common global language to communicate in.

It also beats the fuck out of learning Latin, French, or German.
posted by Blasdelb at 3:32 PM on March 24, 2012


Metafilter: Screw your stupidity; here’s a fact-bomb for you
posted by benzenedream at 3:38 PM on March 24, 2012 [4 favorites]


Pfft, whatever. I love science writing. Science writing seeks only to inject as much information into your head in as few words as possible. Passive voice is appropriate here, because I (supposedly...) don't care about the agent performing the experiment, all I care about is what was done.

When you watch Pimp My Ride, do you want to hear "Alicia may have had an inkling when the phone screeners called her, or certainly when the camera crew arrived, but she could not yet know what we intended to do to her vehicle." No, you want to see the fishbowls and purple plush without delay, right? Hence the stock phraseology. Like Pimp My Ride, science writing is all payoff.
posted by pH Indicating Socks at 3:43 PM on March 24, 2012 [1 favorite]


As a scientist, I'm kind of baffled by some of the points in the article, and some of the points made above. Some ring true, yes -- but others violate what I have been taught is good scientific writing, what I try to do in my own papers, and what I try to teach to my students. The main principle I strive for is clarity. The same is true for many scientists I know.

Specifically:

- I think "lone" is a perfectly cromulent word in that context. It is also okay to write about "showing" as well as "demonstrating". In fact I usually use both words so as to avoid monotony.

- I try not to pad a paper with unnecessary references. It just makes it harder to read and often works against word limits. On rare occasions I've been asked to cite more in the review process, but that was almost always some relevant reference I missed; I can't think of a time I have been told that my reference list was too short in general, for no reason.

- I know some people are taught to use the passive voice, but at least in my field, active voice is far more common, and something I and my collaborators teach to our students. I have never, ever been reproached for writing in the active voice during the review process. In fact, I am often commended for making the writing so clear.

- Of course you always write out the numbers if they are ten or lower!

- Never ever start a sentence with "obviously". In fact, one of the main writing tips I learned as a graduate student was to aggressively trim any words like that out of a sentence wherever it might appear.

- It is okay to use interesting words, or wry turns of phrase, as long as they are clear. Sometimes the clarity is hard to maintain with these, so the bar is higher. But I have even been known to include humorous footnotes with (gasp!) actual jokes in them, and again, they always get published without comment.

- But #12, hah, I totally agree with. :)

My point is not that I'm all that great of a writer - I know I have a lot of tics that I still struggle with, and could improve in a lot of ways. But this article in no way describes all scientific writing. It describes a certain species of bad scientific writing, yes, but if you're being encouraged to write this way, I say, throw off the yoke of oppression! Embrace clarity! Eschew stupid rules! Most reviewers will love you for writing something that is readable, and if anything, it will pay big dividends.
posted by forza at 4:01 PM on March 24, 2012 [5 favorites]


Yay, I have a chance to post this. See also Sand-Jensen (2007):

How to write consistently boring scientific literature
posted by carter at 4:20 PM on March 24, 2012


The other thing to consider: many scientists may be bad writers. By avoiding all flourishes, you're giving them less rope to hang themselves.

Everyone who has ever read a paper or report that I wrote should be thankful for the norms of scientific writing. Left to my own devices it would be just awful.

I promise you that the alternative to dry papers with limited vocabulary is much, much worse.
posted by no regrets, coyote at 4:22 PM on March 24, 2012 [1 favorite]


It is okay to use interesting words, or wry turns of phrase, as long as they are clear.

I think the struggle here is a result of the fact that good writers are underrepresented among the set of all scientists. Because everyone has to a write a paper at some point, someone has to put the work into bringing poor writers up to speed. My guess is that the people in charge of doing so usually hand down straight-forward guidelines to their tutees so that those poor writers can bang out something approaching a well-written paper with the application of a few simple heuristics. The problem happens when those people get tenure and it falls on them to tutor their underlings in the ways of scientific writing; that's the point where useful guidelines become harmful dogma.
posted by invitapriore at 4:32 PM on March 24, 2012 [1 favorite]


For me, as a non-native English speaker in a non-english-speaking country, scientific prose is what I read and write the most, and it shapes my use of English also in other contexts, such as here. If you bothered to look back through my comment history, I'm sure you'd find they read in the same dry way as scientific papers do. Lots of "though", "however" and "hence", with heaps of commas between them. Actually, you don't need to browse back, I think this is quite the meta-comment. However, if the academic language was less constrained, I would be at a greater disadvantage to my english-native peers, so all in all I'm quite happy with how it is.

I came to this thread because I was facing an issue of active vs passive voice, and will face it again as soon as I have clicked "post" and go back to writing my PhD thesis. I'd prefer to describe an experiment in an active voice, but I state explicitly elsewhere that I did it by myself. What do I do then, when "I" is too personal and "We" is too majestic?
posted by springload at 4:49 PM on March 24, 2012


I have a scientist friend who is attempting to launch a blog to "practice" writing. She has the opposite problem that this post describes: she's been writing scientific papers for so long that she would never dream of using a word like "lone."

She sent me the first post and if you can imagine it, it's like a blog abstract. Hello. My name is Science Lady. This is what I will accomplish with my blog. This is the method I will use. You are reading my blog. Good bye.

It's seriously going to be either the most boring blog in the history of the world or a runaway internet success in that people will think it's some kind of joke. Could go either way.
posted by sonika at 4:54 PM on March 24, 2012 [8 favorites]




Re: making sure scientific papers are as easy for non-native English speakers to read and understand, I understand and acknowledge this point. And I most certainly do agree that we shouldn't be potentially screwing over anyone who is a non-native speaker by making things difficult, because many are already at a possible disadvantage or, at the very least, a significant inconvenience.

But I can still lament how terrible so many articles are to read, and how much I absolutely hate writing them in the standard way. The process of writing one of them made me absolutely hate my own project, and maybe I should because maybe it wasn't really worthwhile in the end. But I really hope these other papers don't.

p.s. I already know I'm not cut out for research/publishing in academia in the long-term. The way this sort of thing crushes my soul is only one indication.
posted by six-or-six-thirty at 4:59 PM on March 24, 2012


Science writing seeks only to inject as much information into your head in as few words as possible.

I am a tech writer. I have sometimes been asked to edit scientist's writings, or incorporate them in the manuals I write. In my experience, the "as few words as possible" part of your assertion is not even remotely true.
posted by Kirth Gerson at 5:00 PM on March 24, 2012


The other thing that is bothering me about this paper is that there are already so many ways to massage a paper so that you get a better publication. Selectively citing your data, stretching your results to support some argument that isn't actually supported, using questionable statistical methods, rejecting other people's papers and then replicating the results yourself - these are all things I've witnessed scientists doing to get published in the best possible journal. We don't need to add florid writing to that list. Already, you can sometime read between the lines of obscuring words and realize that someone is trying to pull the wool over your eyes.

I mean, it's this like some of the fact-checking stories lately - sometimes journalists or comics or whomever change the facts to make the story better. In science, you're not supposed to do that. Sure, there are acceptable ways to do it (who doesn't write the introduction last?) but overall, it's not a story, it's science. Just give me the facts, not the 'facts'.

Plus I agree with forza that a lot of these are taught as things not to do. My field is slowly switching to an active voice so it's nice to have choices now.
posted by hydrobatidae at 5:12 PM on March 24, 2012


Oh, well done!

IANAS, but just having read through several high-school level science presentations during our county science fair, I can substantiate the author's observations (see? I can sound scientific, too!).

Scirnce teachers should award extra credit for kids who are able to combine several techniques, for a Science Bingo! Or Yahztee!, like so:

As has been clearly demonstrated 867-5309, 5138008-71077345, # of licks it takes to get to the tootsie roll center previously through several notable studies, including the pioneering work of my esteemed predecessor, Vladimir Putin, ricin is both effective and virtually untraceable when administered even in quite small quantities...

John Doe
Jonathan Smith
John Carter of Mars
Steve "The Human Guinea Pig" Williams
The late Thomas Carruthers III
Marie Curie
Vladimir Putin
Professor Neville Longbottom, PhD, MSW, ASPCA
posted by misha at 5:17 PM on March 24, 2012


Okay, I have an admission. I hate writing. I am currently procrastinating writing two papers (finishing my dissertation) so I'm feeling a lot of pressure already for being a crap writer. And this guy's comments have not helped.

I would ideally just print up a bunch of figures and tell people to figure it out themselves.
posted by hydrobatidae at 5:17 PM on March 24, 2012 [3 favorites]


As someone who translates SCIENCE! into English for a living, I can sympathize. Deeply.
posted by Rangeboy at 5:21 PM on March 24, 2012


Having just finished final checks on my current MS, I will now reflect on my particular process of SCIENCE! writing:
First draft post outline: Write with whatever language I feel like, highly florid vocabulary and airlessly complex sentence structures.
Subsequent self edits: Rueful rewording of the majority of florid vocabulary, somewhat gratifying simplification of sentence structure.
First outside edit: A RUTHLESS PURGE of remaining florid vocabulary. Additional simplification of sentences.
Second to last edit: A begrudging acceptance of most trimmings, but, on my part, a vicious clawback on one or two high octane words that are so exquisitely precise that I can't stand to replace them with more common, or wordier phrases. By this time we're in nickel and diming word count territory, so me and the dictionary often win.
Final Draft: A terse, desiccated shell of the original, but acceptable to all parties involved. I still have a few words remaining that bring me joy just to see them on the page, knowing I fought for those beauties and they've come through all those drafts.
posted by Cold Lurkey at 5:35 PM on March 24, 2012


It's a good thing mathematicians aren't scientists, or I'd take issue with this article's thesis. This person has never read Fruit Salad [pdf] by Gyarfas or The Two Cultures of Mathematics [pdf] by Gowers.

I took a class on hypergraphs with Gyarfas once. It was great! He's also the source of my favorite open problem in graph theory, called the Tree Packing conjecture.

You have two hands.

In your left hand, place an arbitrary sequence of trees, one on two vertices, one on three vertices, and so on, up to a tree on n vertices. Recall that a tree with k vertices has k-1 edges. Then notice that you have a total of (n choose 2) edges in your left hand.

In your right hand, you have the complete graph on n vertices, which, coincidentally, has (n choose 2) vertices.

Question: Can you 'pack' the trees in your left hand into the complete graph? It's proven by computer up to something like n=10... But the problem itself is still very open!
posted by kaibutsu at 5:50 PM on March 24, 2012


Probably the best line I ever got into a paper is:
Heroic efforts to construct a more reasonable curve for mature MS cows were deemed unnecessary because fewer than 3% of them have lactations longer than 500 d, whereas 90% have lactations less than 400 DIM, where the standard curves appear to be quite reasonable.
By the standards of my field that's Shakespeare right there. If you're having trouble sleeping you are welcome to read the entire work free-of-charge in Journal of Dairy Science (Cole, J.B., Null, D.J., and VanRaden, P.M. 2009. Best prediction of yields for long lactations. J. Dairy Sci. 92(4):1796–1810.)
posted by wintermind at 6:05 PM on March 24, 2012 [3 favorites]


The other main culprit for this hegemonic variety of "science writing" is page limits. The best journals (Science, Nature, PNAS, Physical Review Letters) have page limits that are downright draconian. Yes, please sum up the last nine to twelve months of ground-breaking research into four to six pages. With large figures using up much of that space. Resorting to shorthand and removing every bit of unneeded character from your sentences falls out of this emphasis on brevity. Of course, sometimes that five page paper has a fifty page supplementary info that is the real paper, and that will often be written with much more personal style behind it.

The flip side of this is that papers take on a curious role. Actual cutting edge research is discussed in talks (which are culturally allowed to range in personality from cardboard to real, funny person) and the conversations that follow them. Papers are useful for having something to cite and having a reference for a result and its context. But learning about an interesting, highly relevant result from a publication means you're out of touch with the state of the field. I strongly suspect that one reason for this is that the paper itself is reduced to such a partial document that you can't actually communicate what you want to in its fullness and get it by referees and editors.
posted by Schismatic at 6:42 PM on March 24, 2012 [3 favorites]




I recently read a paper where chaldron was used as a unit. I've used acres, roods and perches/rods, myself.

it's less weird when writing history, but we also do tend to define the weirder measurements.
posted by jb at 9:00 PM on March 24, 2012


I remember learning how to do this sort of writing in a high-school science camp one summer at a local university. Just learning how to get useful information out of scientific papers for citations took a significant amount of time, far more than we ever spent in the lab. As for the actual writing part, well, before we got better at it, between three-fourths and one-half (er, 3/4 and 1/2? heh) of what we wrote in our early drafts was ruthlessly slashed and burned by our disgusted faculty adviser. He also made us practice the sound bites and slide transitions in our PowerPoint presentation until they were utterly crisp and dry. We won an award for our little pilot study, so I guess it worked...but I remember it being pretty difficult to go from the type of writing we did in the essays to get accepted to the fact- and citation-laden scientific writing we were doing by the end of the term.
posted by limeonaire at 9:16 PM on March 24, 2012


"Scientific writing" is not supposed to be interesting structurally: the content is supposed to be what's interesting. The goal is to express complex concepts as concisely and unambiguously as possible. Pretty prose that makes you stop and marvel at the author rather than the facts that are trying to be communicated is a failure.

Technical writing is difficult to do well, and when done well is almost invisible.
posted by cardboard at 5:32 AM on March 25, 2012


If the information isn't interesting in itself, you shouldn't bother writing it. The whole point of writing is it because it should be worth writing about, she said, ending in a preposition.
Invisibility is effective in a lot of writing.
posted by provoliminal at 7:35 AM on March 25, 2012


As a minor mid-career scientist, may I contribute my strong feeling that communication and writing are *the* core skills required of a scientist. There is good scientific writing out there. Sir Arthur Eddington is a joy to read. Pinker and Dawkins, for all their faults, read well. If you believe science is about finding the one pure truth, then you will probably have little reason to expect good writing in that cause, but another view is that science is about arriving at the very best story of consensus that we can, and from that point of view, writing is absolutely fundamental.
posted by stonepharisee at 7:50 AM on March 25, 2012


The function of technical science writing isn't about being pretty... it's about getting citations for your paper. Elegance isn't a factor.

Popular science writing has different motives, like capturing the imagination. The 20th century classics like Sagan and Gould can be quite elegant.
posted by ovvl at 10:12 AM on March 25, 2012


The best journals (Science, Nature, PNAS, Physical Review Letters) have page limits that are downright draconian.

Here, "best" corresponds to "most prestigious." There is growing concern, however, that these page limits (among other problems with the editorial priorities of the journals, particularly Nature and Science) are perniciously undercutting the quality of the work itself.

This isn't to say that an article in a top-tier journal is automatically bad, of course. The bar is set high enough that papers with obvious problems still can't make it past peer-review. The difficulty arises when the problems are subtle. In these cases, the terse, summary nature of the main article can easily wallpaper over substantial methodological mistakes that can undermine the plausibility of the conclusions.

The solution? To read that 50-page supplement, and to do so carefully. And, with embarrassing regularity, these reveal questionable methodology, basic statistical mistakes, and logical fallacies. The problem is that editors, reviewers, and the authors themselves simply don't care as much about the quality of the supplement, despite it being mandatory reading for anyone who wants to understand how the conclusions in the main body of the paper were reached.

This became clear to me during an interdisciplinary discussion of this paper in Science, whose main-body-text conclusions are utterly hamstrung by the fine print. On its face, this paper typical of Science as a journal: A headline-baiting conclusion described in vague terms, with a promise of rigor to be had in a lengthy supplement. The trouble is that when you read the supplement, it's hard to find an aspect of the study that wasn't flawed. Their framework for defining political ideology was not only flawed, but obviously failed to conform to the actual opinions of the participants surveyed; measures of skin conductance were collected under less-than-ideal circumstances; double-blind considerations were insufficiently addressed; even the way in which surveys were scored make an amateur-hour mistake.

If you read that supplement carefully, and have the knowhow to see it, you are led to the conclusions that the main argument put forward by the researchers is very likely incorrect: Their results should have been reported as null. And they weren't, because the authors were willing to cut corners and hiding their mistakes in the supplement let them sneak it past editors who were probably thrilled to publish something that got Science mentioned in general-circulation newspapers
posted by belarius at 12:24 PM on March 25, 2012 [1 favorite]


Page limits are there for the publication to be able to bring a variety of stuff to print fast and frequently.

If it's a really good result and one relevant to your work, reading the supplement is a must, in any case.

Anyway, we're getting to the point where people will all have tablets in 5-10 years, so the days of printed publications are numbered (and not page 1, 2, etc.) — and that means the necessity for page limits will evaporate.

Having things digital will make it very easy to have supplemental materials embedded into the main paper, for people who want to explore the research in more detail.
posted by Blazecock Pileon at 3:39 PM on March 25, 2012 [1 favorite]


So I've learned that to answer questions like a scientist, you start your responses with "so," instead of "well" or "uh."
posted by TheSecretDecoderRing at 4:21 PM on March 25, 2012 [3 favorites]


Speaking of "learning Latin, French or German", there are still PhD programs that require learning a language well enough to read a journal article in that language. Although I've heard that computer languages count.
posted by sciencegeek at 5:58 PM on March 25, 2012


My main problem with the supplements in journals like Nat/Sci is that they are the inverses of the actual articles: insanely long (seriously, 50 pages is nothing, I've seen 90-100 page supplements) and completely unconstrained in form, not to mention unproofread, untypeset, and unedited. Nobody is under any pressure to make it concise or even readable.

I understand the economic incentives that have caused papers in Nat/Sci to become shorter and shorter, but I think there's no reason those journals couldn't enforce some reasonable page limits for online supplements as well (at least in my field, ≤5p. supplementary methods and ≤10 supplementary figures/tables total, with normal-length legends, should be plenty).
posted by en forme de poire at 11:27 AM on March 27, 2012


I don't imagine anyone will bother looking these up, but if you want to read examples of excellent, relevant (at the time), well-written research, these are 2 of the better ones in my field. The first presented a serious, novel approach in an entertaining way. The second well and truly nailed the lid down on a wrong idea.

Sheldon, R. W., and S. R. Kerr. 1972. The population density of monsters in Loch Ness. Limnology and Oceanography 17:796-797. (Predictably, some in the field attacked this one for exposing the discipline to ridicule. Humbug.)

Larkin, P. A. 1977. An epitaph for the concept of maximum sustained yield. Transactions of the American Fisheries Society 106:1-11. (A slight reworking of an after-dinner speech. It even uses similes!)

There is absolutely no reason whatever that good science cannot be reported accurately, succinctly, and in an entertaining way.
posted by dmayhood at 6:37 PM on March 27, 2012


« Older A Teacher's Open Letter to Her 8th Grade Students   |   Richard Strauss' "Four Last Songs" sung by Jessye... Newer »


This thread has been archived and is closed to new comments