A strange percentage

From time to time reference is made to a report by a number of scientist asserting
that there is:

"a 90 per cent chance that humans were the main cause of climate change"

I'm not at all sure what that means, mainly because I can't imagine any method by which such a figure could be calculated.

In everyday conversation someone might say "90% certain" to indicate moderate, but not complete confidence in the truth of the proposition in question. However, in scientific discourse as opposed to casual chatter one expects a number to be the result of some calculation. The choice of 90 indicates a reason for thinking it is greater than 89 and less than 91.

I assume that the 90% is meant to be a probability, so the claim is equivalent to:

"The probability that humans are the main cause of climate change = 0.9"

I think that such a use of probability is untenable.

In the 1930's and 1940's philosophers of science, including some scientists in their philosophical moments, tried to solve the notorious Problem of Induction by assigning probabilities to hypotheses. After fierce controversy in which Sir Karl Popper was prominent, those attempts are now generally accepted to have failed. I consider the matter in chapter 6 of my Notes on Philosophy.

I feel a general unease about the discussions of climate change.

It is difficult to distinguish change in climate from fluctuations in weather. Such a distinction would require careful statistical analysis, yet much of the discussion is conducted like the evangelical campaign of a millenarian sect, complete with denunciations of any heretics who dare to express any doubt. That is not the behaviour of honest scientists.

On the other hand the theory of climate change is supported by what I like to call double verification. As well as statistical data that seems to point to some degree of change, there is independent evidence of a mechanism which one would expect to produce such a change, namely increases in the levels of greenhouse gases in the atmosphere.

I wonder how accurately one can predict changes in climate. The weather is chaotic, so it is worth considering the possibility that climate, when it varies, also does so chaotically. However as climate is a sort of average of weather over moderately long period, chaos in weather does not in itself imply chaos in climate.

I believe that something is going on, though I'm not sure precisely what. People would be wise to brace themselves for a shock of some sort, yet to keep an open mind about precisely what form that shock might take.