A row over the apparent wandering ways of one of the universe鈥檚 fundamental constants has been fuelled by claims that a crucial study on one side of the debate suffered from 鈥渂asic flaws鈥�.
The story began in 1998 when John Webb of the University of New South Wales (UNSW) in Sydney, Australia, claimed that alpha, a constant which determines how matter and photons interact, was smaller 12 billion years ago than it is today. His finding was based on measurements of the frequencies at which interstellar gas clouds appeared to absorb light emitted from very bright objects known as quasars when the universe was young.
Variation in this constant over time would have deep implications, and some physicists say it points to the existence of the extra dimensions required by string theory. What鈥檚 more, alpha governs the strength of electromagnetic interactions, the vacuum energy of empty space, and the strength of the weak force. If it was subtly different billions of years ago, we might have to rethink our ideas of how the universe evolved.
Advertisement
In 2004, however, a group led by Hum Chand of the Inter-University Centre for Astronomy and Astrophysics in Pune, India, analysed ancient quasar light captured by the Very Large Telescope at the Paranal Observatory in Chile, and found no variation in alpha.
Now Webb claims that Chand鈥檚 analysis was 鈥渙versimplistic鈥�. In a paper to be published in Physical Review Letters, Webb says Chand鈥檚 group failed to properly take into account certain relativistic effects and Doppler shifts that would have changed the apparent frequencies of quasar light. This would have led to large errors in their calculation of alpha. 鈥淚f you fit the data incorrectly, you get spurious results,鈥� Webb says. 鈥淚t鈥檚 as simple as that.鈥�
Chand and colleague Patrick Petitjean of the Paris Astrophysics Institute in France think Webb鈥檚 conclusions are far too sweeping. In a response in the same journal, they say they have now taken these effects into account, and that although the estimated error in alpha鈥檚 value over time has increased, there is still no strong evidence that alpha was different in the past.
Webb says that the error has now increased beyond the point at which the Chand analysis can be considered stringent.
