A worried family-member sent me a link to this paper, published in Nature, which is supposed to make us scared, once again, of WIFI and mobile-phone emissions. I decided to have a close read, and reported back:

Well, it starts with the statement that “All types of man-made EMFs/EMR – in contrast to natural EMFs/EMR – are polarised”. I suppose I’d better not take for granted that you know what polarisation means. It means light (or any other wave) that’s travelling and aligned at one angle, rather than scattered at many random angles. And it proposes that all “man made” waves are polar, and implies that those made in nature are not.

This is very obviously not true.

For example, the “man made” light coming out of your light-bulb is not polarised, but the EMF emitted after a (natural) lightning strike *is* highly polarised. Also, because of Rayleigh scattering in the atmosphere, the sunlight that reaches us is also substantially polarised (i.e. the atmosphere filters out light at certain angles), not to mention sun that is reflected off any smooth surface, which is completely polarised. If it weren’t, polarised sunglasses wouldn’t work! So, if within the first paragraph, we have manifest untruths, I
do wonder whether it’s worth Fisking the rest of it.

But I did find it amusing to note that in the paper itself, they did admit how puny “manmade” EMF is. Again, quoting DIRECTLY from the paper:
“Solar EMR intensity incident upon a human body ranges normally
between 8 and 24 mW/cm2″
vs
“intensity from a digital mobile phone handset upon a human head
during “talk” emission is normally less than 0.2 mW/cm2″

Also, I love the fact that they have to admit that your immediate family emit greater EMF than any WIFI transmitter:
“radiation from every human body at normal temperature, have
significantly larger incident intensities and exposure durations on
any human than most artificial EMF sources”

Again, this is a direct quote from the paper you sent me 🙂

They admit that this is a problem for their hypothesis, that manmade EMF must, somehow, be dangerous, even though they have no statistical evidence for this. They beautifully beg the question by suggesting they’ll propose to “explain theoretically why, even though man-made EMFs are puny, they’re rendered harmful by being polarised “in contrast to the natural ones”. Firstly, they’re having to invent a theoretical way to explain a problem that has no evidence of existence, and secondly, their “theoretical” explanation has its foundation in a postulate that is empirically false (that there aren’t polarised waves in nature). Oh dear.

But don’t take my word for it. In a throwaway comment, they admit that, actually, yes, we experience plenty of polarised radiation in nature:
“Unpolarized natural light can become partly polarized to a small
average degree after diffraction on atmospheric molecules, or
reflection on water, mirrors, metallic surfaces, etc”. But, hold on, they admitted that our exposure to sunlight is several orders of magnitude greater than our exposure to even close “man-made” EMFs, so even if the “small average degree” of natural polarisation were true, it would still, in absolute terms, be as great as (or greater) than the man-made EMF’s polarisation. In other words, 10% of a million quid is more than 100% of a thousand quid!

I note that the reference for their admitting that natural EMF does, in fact, exhibit polarity, is a textbook. I used Google Books to search that text-book, and was amused to find exercises within asking the reader to calculate the angle of reflection off a lake whereby
100% of the sun would be polarised and so on. So the book discusses blithely a common situation where you’d be exposed to polarised radiation up to 120 times more powerful than holding a mobile phone to your ear. Again, this is just using THEIR reference! And that’s with a mobile phone which is, in turn, about 100 times more powerful than the
average wifi signal.

Thus, if you take this paper seriously, you should be 1200 times more worried about the polarised glare off Lake Superior on a sunny afternoon than in having your infant lie directly on top of your wifi router.

They go on to suggest that the magnetic fields associated with man-made radiation are “are accused [sic] for an association with cancer”. They provide a reference for that claim. Which I follow. It involves gerbils whose brains had been purposefully harmed, and then exposed to a close magnetic field caused by a 50hz circuit. After decapitating them, the researchers did, indeed, notice a difference between these gerbils and the controls. These gerbils had LESS evidence of oxidative stress! Let me quote directly:
“Based on reported results, it is obvious that 7-day exposure to
ELF-MF (50 Hz, 0.5 mT) can reduce oxidative stress in the brain of
gerbils submitted to 10-min global cerebral ischema”. Notice the word “reduce” there. They postulate that the magnetic field gently increases endogenous antioxidant production, which is substantially protective. This is the same gentle hormesis used to justify, say, sun exposure and eating broccoli, and again shows how problematic it is to assume that exposure to anything that might cause bodily disruption is uncomplicatedly harmful. After all, the only reason vegetables are good for us is because, paradoxically, they are bad for us.

So a study that suggests EMF be used therapeutically based on the obvious anti-carcinogenic protective effects they observed on their animals is perhaps not the best to use as a reference for “accusing” an “association with cancer”. This can only mean that the authors of your paper did a pubmed browse for titles, thought this one would be good for them, but didn’t bother actually reading the results which were rather counterproductive to their hypothesis.

The other reference they provide for their scaremongering is a huge summary of all the evidence available on this subject, produced by the World Heath Organisation and accepted with much hoo-ha by the woo brigade. It’s utterly, exhaustively, exhaustingly comprehensive. It is urgently vital that this paper be a valid reference for the worries upon which your paper is based, otherwise the fear-mongerers’ postulated mechanisms have no harmful referent, and are mere discussion of what dance steps those putative angels dancing on pinheads used. In other words, if the WHO huge referenced dataset reveals cause for concern, then a search for a plausible mechanism is indeed a logical pursuit. However, if that dataset reveals no cause for concern, then there is no existing harm for which a mechanism could be found, and the pursuit becomes as quixotic as people who try to fix the unfortunate data of the world into a Young Earth Creationist mindset.

So let’s see what the exhaustive dataset, upon which they base the drive for their whole hypothesising, says about the harm that man-made EMF’s cause.

Let me quote directly:

“There is limited evidence in humans for the carcinogenicity of
extremely low- frequency magnetic fields in relation to childhood
leukaemia.

There is inadequate evidence in humans for the carcinogenicity of
extremely low- frequency magnetic fields in relation to all other cancers.

There is inadequate evidence in humans for the carcinogenicity of
static electric or magnetic fields and extremely low-frequency
electric fields.

There is inadequate evidence in experimental animals for the
carcinogenicity of extremely low-frequency magnetic fields.

No data relevant to the carcinogenicity of static electric or magnetic
fields and extremely low-frequency electric fields in experimental
animals were available.”

The final conclusion is that EMF and ELF electric fields are “not classifiable as to their carcinogenicity to humans” and that any carcinogenity of magnetic fields is only accepted as such if one accepts the “limited evidence” of leukaemia as valid.

So what about the “limited evidence” about childhood leukaemia? Well, the review admits that the evidence is very weak and epidemiological (like sat fat) and may well”be affected by “selection bias”. It goes through the substantial problems with the epidemiological evidence, and concludes “If the observed relationship were causal, the
exposure-associated risk could also be greater than what is reported”. The focus on leukaemia is strong because there was a scare in the 1970s that specifically focussed on it. Studies followed, in ever increasing number, and the cacophony of confounding variables and selection-bias became ever more strident. The paper admits that the best studies, that work hard to gut through this cacophony, find no causal signal at all.

So the dataset that should provide the foundation stone upon which their following hypothesis of the mechanism of harm is built is made of sand. Worse than that, the actual experimental data revealed in their only other reference (the aforementioned gerbil study) suggests that whatever mechanism there is or is not at play is empirically suggestive of productivity rather than harm!

The rest of the paper is lots of maths and magicks, and, built on sand as it is, requires no more serious analysis than any esoterica written by Aleister Crowley. It may be a very pretty hypothesis. The maths might be clever. The incantations internally elegant. But then, so is the theory of the music of the spheres and so forth. So, by all means, enjoy it as a work of aesthetics. But not science.

They conclude, of course, that all modern technology is evil (as is the wont of every Puritan), referencing laughable epidemiology about mobile phones and cancer as they scattergun their conclusions, and bidding thee repent!