Retraction Watch

Tracking retractions as a window into the scientific process

Does irony have a place in science?

with 17 comments

sciengethicsTake us at our word when we tell you this isn’t some exercise in meta-irony or meta-criticism or any other meta-bullshit, but a pair of researchers at Drexel University in Philadelphia have published a paper calling for an end to irony in science.

First, some background: In 2001, an Israeli researcher named Leonard Leibovici wrote a letter to the famously lighthearted Christmas issue of the British Medical Journal describing a randomized controlled trial in which intercessory prayer at a distance — in other words, people praying for other, sick people — was found to improve the health of patients with bloodstream infections. All the more remarkable was that this prayer was “retroactive,” as in, it purportedly occurred years after those sick patients had either left the hospital or died.

The article clearly was a gag (it ran under the heading “Beyond Science”), as Leibovici later admitted — but not before other researchers cited his work, many favorably.

Now, our impulse, and Leibovici’s, as we’ll show in a bit, would be to poke fun at the scientists who took him seriously. After all, who could possibly believe that, absent a time machine, retroactive prayer could affect the outcomes of anything.

But that’s not how Maryam Ronagh and Lawrence Souder see it. Ronagh and Sauder, who work in the Department of Culture and Communications at Drexel, have written a piece for Science and Engineering Ethics titled “The Ethics of Ironic Science in Its Search for Spoof,” which says that articles such as Leibovici’s are a serious threat to the integrity of science with a capital S.

After a lengthy exegesis on the definitions of irony, studded with references to Kierkegaard, Socrates and other philosophers, the authors take aim at Leibovici. They note that no fewer than 15 papers cited his bogus article “at face value,” including a 2009 Cochrane Review and another that called the sham study “the largest, most important, and best-funded research studying consciousness and nonlocality.”

They also note a few critics, one of which was a scholar who wrote, evidently without irony, in a letter to BMJ that Leibovici had failed to obtain informed consent for his work, and that “Ethical issues should not be limited to linear time.”

At this point, we found ourselves wondering if Ronagh and Souder were operating on an even higher level of irony, but they pulled us out of the rabbit hole with the following admission:

It’s worrisome to speculate whether the commentary in BMJ is itself ironic. If so, we fall into a hopeless morass of uncertainty about all content in BMJ, the research corpus, and even this analysis of ironic science itself, a condition Booth (1974) calls unstable irony and a threat to epistemology generally.

Worrisome, indeed! Which is why Ronagh and Souder end their manifesto with a call to arms. Although they admit that Leibovici was trying to point out a weak spot in science and science publishing: to wit, not everything that can be studied can be studied validly. But, they write, his instrument was too dull, or maybe too sharp:

This lesson may be important but its cost is dear, for the means of delivery — irony — has enlightened some but misled others. For this reason the blemish on the scientific record left by Leibovici’s paper must be expunged. Though it will be disappointing for the reader and tedious for the author to issue a retraction (much like the anticlimax that results from explaining a joke), Leibovici’s community will not be secure in trusting the research record otherwise.

We don’t profess to know much about irony (though we have been accused of flinging snark, which is a base form of the art, we suppose), but Ronagh and Souder then stick a toe into waters with which we are a bit more familiar:

Based on the this study, we worry that the integrity of the scientific record is as vulnerable to ironic science as it is to retracted research. … Readers of scientific research will not expect a published paper to express certainty about the truth of its content, but they should expect certainty about the truthful intent of its authors.

First off, it’s simply not true that science is more vulnerable to irony than retractions. As we’ve reported, the number of cases of wool-over-the-eyes spoofs (we’re not talking fabrication of data) is much smaller than the number of any-cause retractions. And even the latter — some 500-600 — is a small percentage of the 1.4 million-odd published papers each year.

But the larger point, that researchers should be able accept what they read at face value and without any skepticism or intellectual effort is akin to saying that you should drink from any bottle you come across because, well, it’s liquid in a bottle, by golly!

Sorry, but if some scientists are too gullible to detect an obvious spoof/hoax, they should probably find a new profession.

On the off chance that we’d fallen for a hoax, we emailed Souder, who said the suggestion:

made me laugh out loud.

Our paper is quite sincere. We first pitched the paper to the Journal of Medical Ethics, which as you may know is owned by the BMJ Publishing Group. They declined it; perhaps they thought is was ironic.

Perhaps.

Although we’d love to take another whack or two at the paper, we figured we’d give Leibovici the last few words. Here’s what he wrote to us:

Their point is naive, and to my mind so naive as to be dangerous. They’re are assuming a strict contract between authors, editors and readers of scientific publications saying everything written should be taken at face value. Irony has no place in scientific publications. They claim that the danger of people mistaking an attempt to change the past (based on a quote from Jorge Luis Borges, testing prayer, published in the Christmas issue of the BMJ) as research instead of a play with a database (intended as irony) is too great and that pieces like that should not be published or at least retracted.

To my mind some ‘research’ (on intercessory prayer or homeopathy) fit Borges’s definition of baroque:” .. that style that deliberately exhausts (or tries to exhaust) its own possibilities (of irony or parody)…”. “RCTs” in intercessory or homeopathy contain themselves all the irony or parody that can be invented, almost nothing left (I’m proud that I’ve found something like that, time travel). Some research articles are a parody though certainly not meant to be so  -and we should remind people of that. So irony should not be excluded from medical publication – it’s a tool to detect intellectual trash.

More takes from Rose Eveleth and Neuroskeptic

Written by amarcus41

December 23rd, 2014 at 9:30 am

Posted in sci eng ethics

Comments
  • Jan Witkowski December 23, 2014 at 10:06 am

    Many years ago I wrote some articles on humor in scientific and medical publications, including “Nothing to Laugh at all: humor in biochemical journals”, TiBS 20: 163-168, 1996 . Humor is not quite the same thing as irony but still a subject to be taken seriously by some.

    Nature was the leader in humor in science journals in the period 1965-1995 and the BMJ among medical journals. On one occasion Nature published humor in its scientific letters pages. This was a letter on ‘The effects of sexual activity on beard growth in man’, published on 30 May, 1970. It and the following correspondence should be required reading for those alone on desert islands.

  • Yurii Chhinenov December 23, 2014 at 10:52 am

    Irony (and humor) certainly has place in science, but perhaps not at the Department of Culture and Communications. Culture and Communications is a very serious business.

  • Neuroskeptic December 23, 2014 at 10:53 am

    As I said in my post about this paper, “I don’t think any paper ought to be retracted just ‘because it’s meant as a joke’. Quite apart from making science less funny, this would introduce a subjective element into science… What matters is not the author’s intentions, but the nature of the trial itself.”

    • Oliver C. Schultheiss December 24, 2014 at 3:22 am

      Subjective element??? In science???

      As a scientist, I was raised on the notion that science is and always should be objective! And now you come along and proclaim “subjective elements”?

      That’s it. I’m leaving science. Right now.

  • Thomas E DeCoursey December 23, 2014 at 10:54 am

    We need more humor in this world, so I support irony and humor in scientific publications. However, I accept the point that readers, especially non-native English speakers, have trouble recognizing and understanding humor and at best, may feel left out. However, humor is so important in its ability to constrain self-importance and pomposity that I still favor its use. However, it should be done in a way that – even if the reader does not get the joke, s/he still understands the content correctly. Finally, the fact that the ironic paper was cited without the citers’ realization of its ironic nature, is NOT a valid argument against the use of irony. It is simply a demonstration that many citations are made based only on the title of the paper. Retracted papers continue to get citations at an alarming rate, despite the fact that if the citer downloaded the original paper, it would have “RETRACTED” in red ink across each page. Bogus citers like these deserve public ridicule.

  • Peter Apps December 23, 2014 at 12:00 pm

    Keep Calm
    Put Irony in the Keywords
    And Carry On !

  • Sylvain Bernès December 23, 2014 at 12:18 pm

    Irony is difficult to handle because misinterpretations may lead to unexpected consequences. I certainly prefer humor (except sick jokes), provided that it is wisely put in context in a way that does not affect the science in the paper.
    For an example, see the abstract of this math paper:
    http://link.springer.com/article/10.1007%2FBF01218452

    (Found on a MathOverflow thread:
    http://mathoverflow.net/questions/22299/what-are-some-examples-of-colorful-language-in-serious-mathematics-papers )

  • anon December 23, 2014 at 12:22 pm

    2 points –

    1) Is what they’re complaining about really irony? It seems to me that it was really just a problem of a humorous article being treated as “real science”. Irony by definition suggests that the opposite of what is being stated is true in some way. The BMJ article is just funny, not ironic. Whatever is called, it makes reading journals fun!

    2) The problem is that stuff journals choose to put in their “front sections” might still get indexed in PubMed or misread by someone who is sent a link to the article without knowing the proper context. The BMJ article is in PubMed – PMID: 11751349

    So, some care does need to be taken about how humorous (or ironic content) is packaged and disseminated. While the BMJ page makes it clear that the article is “beyond science”, on PubMed, it looks like any other piece of BMJ content.

  • oldnuke December 23, 2014 at 1:58 pm

    A lot of people need more irony in their diet.

    • blatnoi December 24, 2014 at 6:05 am

      It’s easy to say that, but that’s what causes them to just put it into cereal in elemental form. You have to remember that it needs to be in ionic form in order to get absorbed by the body.

      • Peter Apps December 25, 2014 at 2:49 am

        You are correct – what this study shows is that the uptake of irony is enhanced by its being complexed with plausible-sounding technical language, and with the irony-technical complex being dispersed into high-quality ingredients.

  • Leonid Schneider December 23, 2014 at 2:05 pm

    I didn’t expect the Spanish Inquisition! indeed, this discussion is turning Pythoesque. Irony as a threat to Science, similar to that of faulty/fraudulent research and subsequent retractions? Did someone lose all sense of proportion?
    Should PhD students again be banned from having private lives (as they were used to), otherwise their lab performance may suffer from such distractions? Is farting in the lab liable to skew ongoing analytic assays? Perhaps Drs Ronagh and Souder think these are all valid concerns, and possible threats to Science.

    • Sylvain Bernès December 23, 2014 at 4:27 pm

      Perhaps the proportion to be applied here is that the BMJ is not Mickey Parade. Or at least I hope.

  • Narad December 23, 2014 at 4:57 pm

    After all, who could possibly believe that, absent a time machine, retroactive prayer could affect the outcomes of anything.

    I take it that you are blissfully unaware of the (second-degree) reiki symbol “Hon Sha Ze Sho Nen” (and, yes, NCCAM has funded reiki trials).

  • Conrad Seitz MD December 23, 2014 at 6:06 pm

    I love humor and irony in science. I am deeply frustrated when people “don’t get it.” I would almost go so far as to suggest a universally understand signal that “the following is not to be taken seriously” except that some people wouldn’t get that either.
    I disagree as to the specific article being ironic because I think that the writers were saying that the opposite of their conclusions is the truth–that is, it is completely impossible for postmortem prayer to affect a patient’s lifespan. That’s irony, right?
    There is, of course, a dedicated forum for “funny” papers: the Journal of Irreproducible Results. My favorite

  • herr doktor bimler December 23, 2014 at 6:39 pm

    a scholar who wrote… in a letter to BMJ that Leibovici had failed to obtain informed consent for his work

    It’s OK though, as long as he obtains consent at some time in the future, and makes it retroactive.

  • Michael Wise January 7, 2015 at 3:42 am

    Another example is an article on the (entirely fictitious) Australian creature the drop bear that appeared in Australian Geographer. Janssen, Volker (2012). “Indirect Tracking of Drop Bears Using GNSS Technology”. Australian Geographer 43 (4): 445. doi:10.1080/00049182.2012.731307

    Last I looked it had a) had more reads that several year’s worth of “normal” articles put together b) been cited once, which suggests to me that most people who come to read the paper are getting the joke.

  • Post a comment

    Threaded commenting powered by interconnect/it code.