How to write consistently boring scientific literature

As I sit here, trying to write a paper, I found this article entitled “How to write consistently boring scientific literature” very interesting. (via The Annals of Improbably Research”

I’m afraid it’s behind a paywall, so I’ll summarize their findings.

Here’s their table that summarizes their findings in bullet-points.

i-dceabc169673334667317fd7fad1d06f-Boring.jpg

Avoid Focus

This is my favorite one:

If an author really wants to make sure that the reader looses interest, I recommend that he/she does not introduce the ideas and main findings straightaway, but instead hide them at the end of a lengthy narrative. The technique can be refined by putting the same emphasis on what is unimportant or marginally important as on what is really important to make certain that the writing creates the proper hypnotic effect which will put the reader to sleep.

This is unfortunately true. I guess the motivation is to try to avoid making strong statements that might come back to haunt you, but the fear of making a mistake often results in roundabout reasoning and really boring statements.

Avoid originality and personality

Yep, I totally see that one. God forbid you actually say something funny or include some kind of joke. And it usually isn’t your reviewers that will complain about such a thing, but your peers who worry that they will detract from the seriousness of your report.

Remove most implications and every speculation

This almost seems like a repeat of avoiding focus, but the example they provide of what you should do is excellent.

“It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material”
James Watson and Francis Crick (1953)

Now, that strong statement could have been really bad if their paper hadn’t revolutionized the field or proven incorrect.

Leave out illustrations, particularly good ones

I wish our school or departments had regular contracts with scientific illustrators. Some schools do, but there really should be a position at each school where they have a scientifically-literate artist that helps authors conceptualize their findings with clear and pretty art. Ideally, most papers would end with a nice illustration of the changes to the model, or proposed mechanism that the paper discusses. And hey, it’s an extra figure!

Omit necessary steps of reasoning

Pretty straightforward.

Use many abbreviations and technical terms

The worst of science! Jargonization is almost always a sign of incomplete understanding or the desire to hide the fact that your research is really very simple. It reminds me of sociology classes in undergrad, which mostly consisted of teaching the non-sociology students how to translate paragraphs of convoluted language into statements like, “big buildings reflect wealth”, or “poor people have different problems than rich people”. Ugh. I hated that class.

Suppress humor and flowery language

I think this fits with the avoid originality and personality. Also, when one is constantly struggling to stay under page limits, any sentence that can be shortened into something that is terse and boring, is inevitably made terse and boring.

Degrade species and biology to statistical elements

I admit, I didn’t get this one as much.

A very special beech forest, located 120 km away, houses numerous rare plant species. There is no reason to make a fuss about this particular forest because the number of common species in a nearby forest is not significantly different.

Our scientific writing in biology should reduce all species to numbers and statistical elements without considering any interesting biological aspects of adaptation, behavior and evolution. The primary goal of ecological study should be the statistical testing of different models. This is especially true because, on further examination, these models are often indistinguishable from each other, and many have no biological meaning. Hence, writing about them will inevitably produce dry, humorless, uninspired text.

I guess they are discussing the tendency to avoid description of ones samples with any affection?

Quote numerous papers for self-evident statements

This one might be the most difficult to avoid. You have to worry about whether your reviewer will be pissed because you don’t cite their article, or whether you will cite one article but another might have proceeded or predicted the result.

Now if I can just avoid doing these things.

Kaj Sand-Jensen (2007), How to write consistently boring scientific literature, Oikos 116 (5), 723-727.
doi:10.1111/j.2007.0030-1299.15674.x

* I see that Omni Brain covered this article as well. I guess we both like AIR.


Comments

4 responses to “How to write consistently boring scientific literature”

  1. “Omit necessary steps of reasoning

    Pretty straightforward.”

    I hate you. 🙂

  2. Kapitano

    I think there’s a different set of rules waiting tobe written about how to write bad popular science, something like:

    * Use exclamation marks at least three times on each page.
    “…and that’s a lot of nothing!”
    “…black holes fulful all the requirements for being the location of hell!”
    “…a billion billion billion billion billion billion billion times! Wow!”

    * Use units that sound like they give you a feel of scale, but don’t.
    “…enough water to fill three hundred football stadiums.”
    “If the earth were the size of the solar system, a string would be the size of a tree.”
    “A million years – that’s ten thousand grandfather’s lifetimes.”

    * Try to make scientists interesting by making their lives sensational.

    “Alan Turing (who may or may not have committed suicide by swallowing cyanide because he couldn’t live with being gay)…”
    “Einstein fled Hitler’s Germany as a Jew to the safety of tolerant America.”
    “Kurt Godel starved himself to death because he became paranoid that everyone was trying to poison him. However, several years before this…”
    “The great mathematician Richard Montague, shortly before his violent death at the hands of a crazed prostitute…”

  3. Degrade species and biology to statistical elements

    No, this is a real problem, especially in plant ecology. Really interesting and cool plant communities end up as nothing more than indecipherable ordination diagrams. While it’s all the same for a community where you don’t know any of the species, it’s really frustrating when you know the species and can picture them…and all you get is diversity indicies.

    Of course, the specific example of

    The primary goal of ecological study should be the statistical testing of different models. This is especially true because, on further examination, these models are often indistinguishable from each other, and many have no biological meaning

    isn’t just bad writing, it’s bad use of statistics. When you take two obviously different plant communities and test a trivial null hypothesis (that they are not different), it’s uninspired science. When you go a step further and use a test that’s overwhelmed by the 20 common weeds that both sites share, and conclude that there’s no difference between the sites…it’s bad science (but really useful for the developer who wants to show that the patch of forest they want to develop is “no different from any of the other patches of forest”).

  4. Tony L

    Actually, step 6 is interesting. Sand-Jensen starts entry #6 with this great quote:

    “I once knew a man from New Zealand who did not have a single tooth left in his mouth. Nonetheless, I have never met anyone like him that could play the drums”
    Freely after Mark Twain, journalist.

    Now, I think this is something a certain breed of denialist does a lot. You know the type, they’ve got a lot of technical knowledge, and they use it to confuse their unsuspecting and untrained audience. Skip a few steps (or better, invent a few steps that you can then skip) and let confirmation bias do all the work.

    Damn, it’s so easy!

Leave a Reply

Your email address will not be published. Required fields are marked *