Speed of light, redux

Neutrinos probably can't move faster than light after all; somebody made a measurement error. But don't laugh. This is exactly how the scientific method is supposed to work.

Was that a sigh of relief, amusement or disappointment? The world's physicists must have felt one of those emotions last week when they learnt that neutrinos probably do not, after all, travel faster than light.

In theory, nothing can move faster than light. So when an experimental group known as Opera reported last September that it had tracked some neutrinos - tiny, electrically neutral specks of matter - breaking the cosmic speed limit, it appeared that it was "back to the drawing board" for the theory of relativity.

But now one of the Opera partners, the European Organization for Nuclear Research known as CERN, has announced that the faster-than-light finding was probably "an artefact of measurement" - or, stripped of jargon, a mistake. They'll keep repeating the test until they're sure.

Don't laugh too hard. This is exactly the scientific method: form a theory, find a way to test it, do the test, repeat to make sure the result is accurate, confirm or reconfigure your theory, and move on to the next issue.

With intellectual forebears including the 11th-century Syrian scholar Ibn Al Haytham, the scientific method has been - arguably, of course - the principal tool by which humankind has tamed our surroundings.

Modern science, such as subatomic physics, can be hard, costly and slow. But it keeps improving our understanding, one mistake at a time.