I’d like to share something that popped into my inbox this afternoon.

It is entitled ‘Convention on the use of SI units in the Earth Sciences’ and recommends (nay, urges) the use of the abbreviations ‘Ma’ (mega-annum) instead of ‘Myr’ (mega-year) for ‘millions of years’, and likewise ‘Ga’ (giga-year) instead of ‘Gyr’ for ‘billions of years’.

It comes from the several-months-long deliberations of the International Union of Pure and Applied Chemistry (IUPAC), and the International Union of Geological Sciences (IUGS), and they have spoken.

The reaction of colleagues to whom I circulated this, purely for information, ranged from raspberries (Bronx Cheers, for those in the Colonies) to suspicions that I had sent it as a wind-up.

After some thought, my reaction was that this missive represents the latest hangover from the Battle of Hastings (see my earlier post ‘Found In Translation’), and is thus a blatant case of gratuitous Norman French Imperialism. ‘Annum’ is Latin, of course, but ‘year’ is honest, horny-handed Old English (first recorded use in writing: Aethelwold’s Rule of St Benedict, CE 960).

The sentence in the recommendation that struck me most was this: The SI unit of time, the second (s), is impractical for earth scientists and nuclear physicists alike.

The humble second is not alone in its inappropriateness of scale. It is, however, at least moderately useful at the scales in which we live our lives, if not for the measurement of geological time.

How unhandy, however, is the pascal (Pa), the SI unit of pressure, which is 1 newton per square metre, which is similar to being hit over the head by a small chicken (unfrozen) — and even then, glancingly — and is thus no use at all to most people. Atmospheric physicists have to multiply the piddling pascal hugely, as it takes almost 100,000 of these pathetic units to make an atmosphere, and many millions of more to express the kinds of pressures in rocks, stars and so on, required to squash things sufficiently hard to be interesting.

Of comparable gracelessness is the tesla, the SI unit of magnetic flux density, defined as 1 weber per square metre, or for those, like me, who confuse their webers with their henrys, as 1 newton-second per coulomb-metre, or even 1 newton per ampere-metre.

The point is that the tesla, like many efforts at international standardization, is almost (if not quite) as supremely useless as the pascal — albeit in the opposite direction. The problem with the tesla is that it is just too enormous for everyday use. The flux density of the Earth’s magnetic field is measurable in micro teslas. Only with the invention of technology do teslas become useful. So, to be fair, the magnets of very large loudspeakers have flux densities of around a tesla (which seems reasonable), but beware! A modern ceramic magnet with a strength of 1.25 tesla can erase your credit cards, which seems good reason enough to steer clear of one. Medical MMR machines can rack up a few teslas, and after that, matters bcome more esoteric: your everyday, common or garden neutron star generates molar-mangling megateslas. Advisory: most dentists don’t become astrophysicists.

One should, if possible, find units of measurement fitting to the system you’re studying, without having to resort to the SI system of decimal multipliers and dividers, with their rampant panoply of Petas, Exas, femtos and attos, as po-facedly ridiculous as any Pan-European Directive on the Harmonization of the Density of Bristles in Toothbrushes and such like. In any case, these prefixes offer themselves too readily as targets for parody.

For example, biophysicist Walter Gratzer, (reviewing The Dent Dictionary of Measurement in Nature 372, 285-286, 1994) notes:

‘I cannot remember which physicist it was who suggested that zetto may have originated as a misprint for zeppo, after the forgotten Marx brother, and that the zettomole should therefore be followed by the chicomole, the harpomole and the grouchomole. (As the last corresponds to less than one molecule, it could also perhaps be termed the benveniste).’

No, to devise an appropriate unit of measurement is to acquire both elegance and decorum. Consider, if you will, the conundrum faced by a schoolfriend of mine charged with coming up with suitable measurements for snail-racing. Wisely, he eschewed the SI system altogether and devised the FF — Furlongs per Fortnight.

Truly, once we throw off the tyrannical shackles of the Systeme International (Norman French — _naturellement_) we can gambol in all sorts of wild and possibly eccentric pastures.

My favourite non-SI unit is the scoville, which records the relative hotness of chili peppers. Whereas the ordinary bell pepper rates zero scovilles, a jalapeno is rated at between 2,500 and 8,000; a habanero at 100,000-350,000, standard US-grade pepper spray at 2,000,000-5,300,000, and pure capsaicin at between 15,000,000 and 16,000,000.

Note that my sauce (sorry, I meant to type source) has no truck with those effete, SI-style multipliers. No Megascovilles and Petascovilles here: instead we get the beefy, raw power of zeroes, and lots of them.

And this, my friends, is a good thing. For one of the joys of science is the sheer wonder at the magnitude (and, if I may, microtude) of the world and cosmos around us. So if you want a unit, make it fitting. And if there are lots of them — well, simple multiplication gives a much better sense of scale. But exponential multipliers are for wimps.