
Feedback is New Scientist’s popular sideways look at the latest science and technology news. You can submit items you believe may amuse readers to Feedback by emailing feedback@newscientist.com
New kind of microscope?
Science is one of the most fruitful sources of new terminology. There’s nothing like a surfeit of terms like “mitochondrial synthesis” and “quantum fluctuations” to make your writing sound authoritative
Recently there has been a spate of scientific papers containing the phrase “vegetative electron microscopy/microscope“. The term suggests a device for scanning broccoli, but it is utter nonsense. There are scanning electron microscopes and tunnelling electron microscopes, but not vegetative electron microscopes.
One possible explanation was proposed by Alexander Magazinov, a software engineer who moonlights as a watchdog for scientific publishing. He pointed to a 1959 article in Bacteriological Reviews, the text of which was formatted into two columns. Towards the bottom of page 4, the words “vegetative” and “electron microscopy” appear next to each other, in the left and right columns. Old papers have often been scanned using optical character recognition, but such software sometimes struggles to deal with complicated formats. “Vegetative electron microscopy“, according to Magazinov, is “an artefact of text processing”.
However, the journalists at Retraction Watch spotted another possibility, which had been flagged on Reddit. In Farsi, the phrases “scanning electron microscope” and “vegetative electron microscope” sound extremely similar and, crucially, they use near-identical characters: the only difference is a single dot, a diacritic known as a nuqta. This means a tiny mistake in translating a paper from Farsi to English would suffice to create “vegetative electron microscopy”.
These explanations aren’t mutually exclusive, and Feedback is satisfied that we can account for the emergence of this phrase. The bigger question is why it persists in published studies. Are these papers not rigorously peer-reviewed and checked, to ensure a high degree of accuracy and thus preserve the integrity of the scientific literature? Perhaps such “tortured phrases” should be included in a checklist of warning signs that a paper may be plagiarised or fraudulent.
Readers who have encountered similar tortured phrases in their perusals of the technical literature are invited to submit them to the usual address.
A nun too far
Sometimes, Feedback receives a story that feels too good to be true. The set-up is so neat, and the payoff so simultaneously surprising and inevitable, that we doubt ourselves. Is reality ever so neat? And then we remember that the Titanic was the largest ship ever at the time it was built and on its maiden voyage when the bad thing happened. Sometimes, reality is melodramatic. So, maybe we believe this story happened exactly as described, and maybe we don’t.
It comes to us from Charlie Wartnaby, whose late father John was a curator at the Science Museum in London. It relates, inevitably, to the Scunthorpe problem: the difficulty of banning offensive words in online discussions when the same letter strings can appear in harmless words like “peacock” and “Sussex”.
John’s story isn’t, strictly speaking, an example of the Scunthorpe problem, but it’s definitely adjacent to it. As Charlie explains: “In the earliest days of the computing gallery, a machine was set up such that members of the public could type and see their words on a large screen, a great novelty for its day.”
This may seem like an invitation to misbehave. Readers will thus be pleased to learn that staff anticipated the inevitable attempt to write torrents of filth on the large screen for all to see. They drew up “a long list of profanities”, all of which were blocked.
“All was well”, Charlie says, until the system was taken down by the most dangerous person possible: a computer expert. Trying to use the machine, he noticed that some keystrokes didn’t do anything. “Investigating, he managed to pull up the entire list of offending (or offensive) words on the big screen for all to see – allegedly including a visiting party of convent school children and supervising nuns.”
Feedback is prepared to believe 90 per cent of this story, but in the absence of independent verification, we draw the line at the nuns. However, we are also willing to be wrong about this. If any convent school children were in the Science Museum on that fateful day – and we suspect you’d remember – please get in touch.
Yodel-eh-oh
Senior news editor Sophie Bushwick draws our attention to a press release titled “Monkeys are world’s best yodellers – new research”. It describes a study that looks at “special anatomical structures” in the throats of apes and monkeys, called vocal membranes. These membranes allow the monkeys to perform “the same rapid transitions in frequency heard in Alpine yodelling”, but over “a much wider frequency range”, sometimes “exceeding three musical octaves”.
After a build-up like that, Feedback went, with bated breath, to find the accompanying audio recording of a tufted capuchin monkey. We anticipated an ululating call that evoked The Sound of Music or Dutch rock-yodellers Focus. What we got was, approximately, “skroark rark eek”. And now we understand why Sophie told us that she “can’t stop laughing”.
However, a closer look reveals a missed opportunity. By all means, show us a tufted capuchin “yodelling”, but the study also included howler monkeys.
Got a story for Feedback?
You can send stories to Feedback by email at feedback@newscientist.com. Please include your home address. This week’s and past Feedbacks can be seen on our website.