Infinite in All Directions: Why Scientists Should Be Political

Infinite in All Directions is The Wire‘s science newsletter. Click here to subscribe and receive a digest of sciencey curiosities from around the web, on Mondays at 10 am.

Credit: bones64/pixabay

Credit: bones64/pixabay

p-values in history

A two-week old pre-print in the journal PeerJ discusses the history of p-value interpretations in the scientific literature. According to the authors, the modern tendency to overestimate the importance of p-values can be “can be traced back to historical disputes among the founders of modern statistics”. Be warned: it’s less a sequence-of-events story and more a recounting of the evolution of statistical techniques. Nonetheless, those really interested in how p-values have come to cause so much confusion will find the paper useful.


Fusion update

Has another company beat the International Thermonuclear Experimental Reactor to creating the first plasma in a tokamak? Science reports so:

The UK-based company Tokamak Energy has created the first plasma in its ST40 tokamak reactor. The firm will now complete the commissioning and installation of a full set of magnetic coils for the device, which will provide greater control over the plasma. The company plans to achieve a plasma temperature of 15 million degrees by autumn 2017 and have the plasma at 100 million degrees in 2018. At this temperature it should be possible for hydrogen nuclei in the plasma to fuse together, releasing large amounts of energy. Tokamak Energy has ambitious plans to create a fusion reactor capable of generating electricity by 2025 and have a commercially viable source of fusion power by 2030. Unlike the much larger ITER tokamak fusion reactor that is being built in France, the ST40 is a compact device that can run at a much higher plasma pressure. This, according to Tokamak Energy, should make more efficient at achieving fusion.

Let’s still be clear – as we have been with the inertial containment fusion experiment at the US National Ignition Facility: from plasma to fusion is still going to be a billion-dollar engineering nightmare.


Poverty as disease

The sort of thing we’ve all suspected is the case and for which science might finally have conclusive evidence: poverty might be like a disease – in that being poor can cause epigenetic changes in the body that affect our biological wellbeing. Christian Cooper in Nautilus:

The science of the biological effects of the stresses of poverty is in its early stages. Still, it has presented us with multiple mechanisms through which such effects could happen, and many of these admit an inheritable component. If a pregnant woman, for example, is exposed to the stresses of poverty, her fetus and that fetus’ gametes can both be affected, extending the effects of poverty to at least her grandchildren. And it could go further.

I remember reading elsewhere (not able to find the link now) that poverty and mental illnesses may be directly related simply because poor people are more likely to have their lives affected by smaller problems than are those better off. One obvious reason this is the case is that those better off are more able to find short-term alternatives for lapsed services.


Shortcut to sex

An article in Psychology Today details a curious study, which found that after men look at pictures of sexy women, their moral compass goes haywire. The relevant neurological pathway is that women cause men to lose self-control. The study’s author told Psychology Today, “Given that dishonesty can serve as a low-cost and convenient shortcut to acquire resources, power, status, and reputation, men with a heightened mating motive may engage in dishonest behaviours to display preferred characteristics to women in order to promote mate attraction.”

There is an issue with the article’s headline: Sexy Women Sway Men to Do Bad Things, which makes it sound like the women are responsible for the men’s consequent bad choices/actions. Not cool.


Three-pass evaluation

This paper on how to read a paper for researchers throws some light on what reviewers/editors are required to do. S. Keshav, the author, a computer scientist at the University of Waterloo, recommends a ‘three-pass method’, with passes designed such that, when taken together, they allow the reader to assimilate the paper in a critical way. The first pass provides a bird’s-eye view and allows the reader to decide for more passes are necessary. The second pass reveals the paper’s major claims. The final, third pass forces the reader to re-perform the experiment the paper’s authors have conducted (or a solution they’ve attempted, etc.) but knowing only the inputs and outputs, not the logical mechanism in between.

Keshav expects the entire process will take a few hours for experienced readers – i.e. reviewers who are familiar with the subject. For science journalists, this is no good; on some days, I’ve to go through three or four papers in a few hours. And I’ve found the quickest way to deal with them is to simply shoot off an email to a scientist whom you trust as well as someone who is familiar with the contents of the paper. Ask them what they think, what you should be looking out for, whether it’s really significant, if it has any loopholes, etc. Saves you loads of time and also gets you off on the right foot. Scientists are also happy that you’re checking. 🙂


Politicising science

Quartz article discusses the demerits of politicising movements by scientists, with its thrust being that doing so alienates people who fall elsewhere on the political spectrum and who’d probably decided to join you for apolitical reasons. It also suggests that scientists need all the support they can get, so it might be wiser for them to leave the politics out. I’m not sure I agreed but I don’t have to defend myself – Robinson Meyer over at The Atlantic (Quartz‘s parent org) did that very well.

While there is a real risk of alienation when you politicise science, Meyer argues that it’s more important for scientists to stop worrying about who they’re pissing off and actively ensure their self-preservation. And this won’t happen without political involvement: unless you take sides, you’re betting on politicians – effectively, other persons who represent many interests, not just yours – to see sense in supporting you, to perceive lasting value in your endorsement. That can be a lot to ask in times as fraught as these.


I’d like to recommend two blogs for readers to follow.

Suvrat Kher writes Rapid Uplift, a geology blog that can be on the slightly technical side for the casual reader. However, there are often many nuggets on the other side – especially when Kher breaks down something that’s recently been in the news.

Rachana Reddy writes on her blog about space policy, space history and New Space. Her subject can seem daunting at first but read her regularly and you’re going to feel a lot more comfortable, especially about the commerce of space.


From The Wire

Let’s talk (right) about Eman Ahmed

India has an ammonia problem but no policy to deal with it

The marsh of failures that the India-based Neutrino Observatory is stranded in

What made lakhs of scientists march for science across the globe?

An inquiry into common misconceptions about theory and fact

Stories from Renuka Valley as it languishes in the shadow of an unbuilt dam

Do we finally have an anti-ageing drug? No, it’s not that simple.

ISRO is not going to mine the Moon for helium-3


Other bits of interestingness

  1. A small piece of history suggests we should take Elon Musk’s brain-computer interface idea seriously
  2. Is this the best anagram in the English language?
  3. The Chan-Zuckerberg Initiative is batting for bioRxiv
  4. Jigsaw puzzles: a good, simple metaphor through which to understand why more knowledge makes us more polarised, not less
  5. Evidence, and Saganian wisdom, would suggest humans aren’t all that significant in this universe – but astrophysicists have reason to doubt this
  6. Mindstorms is available to read online for free

If you enjoyed the newsletter, please share it with a friend. They can look at previous issues here and subscribe here.