One of the fundamental assumptions of modern physics theory is that it works the same everywhere in the Universe. I always wondered why this should be so. I mean, it makes possible a lot of the calculations used in astronomy, but that just makes the assumption convenient -it doesn’t make it necessarily true. I occasionally amuse myself with the game of “what if it were not true.” As such, I’m always on the lookout for stuff which indicates maybe it isn’t true. Here I present two potential pieces of evidence that the assumption about homogeneity of physical law is not true.

The first one is the oldest one, for which there is the most evidence. It is the Pioneer anomaly. Pioneer was a pair of spacecraft that produced the first up close photos and telemetry from Jupiter and Saturn in the early 1970s. They’ve since left the solar system. They are the farthest man-made objects available for study, so people study them very carefully. In fact, they found a very small, but unexplainable anomaly in their motion. It corresponds to an acceleration which is 10^-10m/s^2 towards the sun. Various explanations have been fielded; some pedestrian, some far out -none are particularly satisfying. Pedestrian explanations: perhaps there is lots of dirt slowing them down; perhaps some large mass is out there influencing the things; perhaps the waste heat is large enough to provide this much acceleration. For non-pedestrian explanations, perhaps gravity works differently than we think. This would be a disaster for astronomers and cosmologists. Perhaps physics theory itself is different in different regions of space. Whatever it is, this effect has been pondered for half a decade now, and they haven’t been able to make it go away. I had the germs of a paper trying to tie this to the solar neutrino problem, but I guess I have more self respect than most cosmologists, so I never bothered publishing it.

The latest one is even more disturbing. It indicates that radioactive decay changes over time. In fact, the rate of beta decay of certain nucleotides seems to be strongly correlated with earth’s distance to the sun. Now, in my opinion, this is rather indicative of systematic error -perhaps detector efficiencies are correlated to something that comes from the sun, like heat. They’re postulating something more radical: that neutrino flux (which varies with distance to the sun by 1/R^2) might have an interaction with these isotopes, or even more radical, that the sun emits some previously unknown field that alters the local fine structure constant. The neutrino instance would be a big deal; lots of nuclear models would have to be revisited. A field that alters the fine structure constant would be cataclysmic. They’ve already looked for and not found the effect in the radioactive power plants of the Cassini probe, which to my mind indicated that the effect is likely systematic error. I’m also not encouraged by the fact that the original paper contains statistical measures which look suspiciously like they did their statistics incorrectly. If anyone came to me with 35 data points and claimed correlation with 4×10^-12 probability, I would laugh at them. I’m guessing they got their “formal probability” by cranking it through some Student-T formula, which is pretty much wrong for calculating correlation probabilities like that. In another case, they actually have the kidney to quote a 2^10-246 probability, which is simply absurd. I mean, that sort of number verges on quantum mechanical absurdity (as in, I think it violates the uncertainty principle, though I don’t feel like doing the math). I suspect such primitive statistical ideology may be common in physics theory; physicists never learn real statistics. Financial analysts are arguably a lot more careful with their statistics, and they’re certainly more sophisticated than most physicists are. None the less, the signals are obviously correlated even if the authors arguably didn’t do the calculation right, and it’s an interesting result.