Wednesday, April 19, 2017

Scientific Consensus Is Untrustworthy.

Gerald Tinsley, Boston Spa.
 Yorkshire Post.

THERE is a very interesting report (The Yorkshire Post, April 11) of large areas of permafrost being thawed by warmer temperatures, and releasing quantities of carbon dioxide and methane. This is evidence that warmer global temperatures are likely to cause increases in these gases in the atmosphere. The “scientific consensus”, however, is that increase in these gases is the cause of the warming. A number of “scientists” found some (rather questionable) data where they were able to show a mathematical correlation between historical indicators of the carbon dioxide content in the Earth’s atmosphere and indicators of global temperatures over the same period of time. From this they concluded that increasing levels of carbon dioxide were the cause of increases in temperature. In concluding this, they fell into a well-known statistical trap. If we show a mathematical correlation between two sets of data (A and B), it does not prove that A causes B. There are three other possibilities, all equally likely, namely that B causes A, that A and B are both caused by some other factor, or that the correlation is purely coincidental. In any rigorous science such a correlation is regarded as an interesting lead to look for further evidence to confirm or rule out any such connection. The report of thawing permafrost and the release of gases strongly suggests to me, however, that in this case B might cause A, meaning that the rising temperature is the cause of increasing levels of carbon dioxide in the atmosphere, rather than the reverse. There is good evidence that the global climate is changing, but it always has been changing, and probably always will. There are very many possible causes, and we know very little about many of them: for example, how much heat is generated in the molten core of the Earth, and how much does it vary? As Galileo demonstrated in 1590, we should be very wary of any “scientific consensus”. Science does not work by consensus and there are many past cases where the consensus has been proved wrong.
Read more at: