The position of requiring a proof on this subject would probably lead to no action ever being taken to curb greenhouse gas emissions. Due to the potential effects of human-induced climate change, principally loss of species and the poorest of the poor being unfairly hurt, I think decisions need to be made based upon the best available evidence.
Attribution studies provide evidence for human-induced climate change. The signal (human-induced climate change) is stronger than the background noise (natural climate variability). Climate scientists like Profs. Richard Lindzen and Roger Pielke Sr, argue that the climate models used in such studies are an inadequate simulation of natural climate variability. They argue that the observed changes in the global mean temperature over the past 50 years could be due to natural climate variability. This is a minority view in the scientific community*.
Consideration of climate sensitivity (the equilibrium global mean temperature response for a doubling in atmospheric carbon dioxide concentration) means that even without attribution studies, most scientists think that the Earth will warm with increasing atmospheric carbon dioxide concentrations. Professors Lindzen and Pielke Sr. have argued that the climate is relatively insensitive to increases in atmospheric carbon dioxide concentrations. Professor Pielke Sr's estimate for climate sensitivity (0.3 deg celsius) is used on the 'Junk Science' website.
The IPCC best-estimate for climate sensitivity is 3 deg C (5 deg Fahrenheit):
'...I think that that 5 degree estimate has a lot of backing for it. It is not based simply on just throwing it into the models and seeing what the models do, but one can do simple, ‘‘back of the envelope’’ calculations with the basic physics in those models that says that if you assume that water in the atmosphere is going to behave in that warmer world the way it does today, that we are going to have relative humidities and cloud amounts like we have today, then that is the number you are going to get, something like 5 degrees Fahrenheit.
You can make that number different if you want. You can assume that the atmosphere is going to get dryer, that clouds are going to shrink. You can make it bigger by assuming the opposite kinds of changes.
To be frank, we do not know whether they might go one way or the other, but in the absence of a real clear understanding of how they are going to change, it would seem like the most conservative assumption would be that they are going to behave much like they do now. So, that is where that 5 degrees comes from.
It is also backed by the kind of sensitivity that we would need to explain the temperature changes that the ice core records tell us happened in connection with the ice ages and the ratio of those temperature changes to the changes in solar energy.
So, I guess I would attach that same 90 percent kind of confidence to that number but with full admission that it could turn out to be too high or too low. But it is the best we can give you right now.'
Climate scientist Prof. Eric Barron. Evidence to the US Senate Committee on Energy and Natural Resources. June 2001. Pages 21-22.
http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=107_senate_hearings&docid=f:76302.pdf
Even if you take the view that Professors Lindzen and Pielke Sr. are right, you must acknowledge that the views of other scientists have value as well. In other words, Professor Lindzen's and Pielke Sr's views are no more provable than the views of other scientists. It therefore seems reasonable to me to look at the potential consequences of a business-as-usual approach, or one that sees us actively trying to reduce our emissions.
* 'The IPCC’s conclusion that most of the observed warming of the last 50 years is likely to have been due to the increase in greenhouse gas concentrations accurately reflects the current thinking of the scientific community on this issue.'
'Climate Change Science: An Analysis of Some Key Questions'. National Research Council. Page 3.
http://www.gcrio.org/OnLnDoc/pdf/ClimateChangeScience.pdf