The practical decision warranted by prudential reasoning that is based on the weight of scientific evidence is therefore that we should immediately begin a process of CO2 reduction. We have seen that there is little or no evidence of a real possibility of catastrophic outcomes of CO2 reduction, so that we do not need to take dissident scientific opinion into account in order to estimate whether we should be cautious about the cost of reduction of CO2 emissions from their ‘business as usual' path. This conclusion is reinforced by the prospect that the cost of greenhouse gas reduction is unlikely to be great initially, giving scientists time to better establish the science on which future estimates of the costs of global warming and of CO2 reduction are based. Moreover, fears of even the long term costs of CO2 reduction may well be exaggerated, since progress with energy efficiency and development of cheaper solar power, carbon capture, and geothermal energy technologies may be greater than assumed in studies like the ABARE research report. Policy makers therefore have every reason to begin now with practical steps to reduce CO2 emissions. The legitimacy of theoretical doubts about the global warming and the effects of CO2 reduction do not affect this conclusion at all. Still less do policy makerd need to concern themselves with Cartesian or Humean philosophical doubts on these matters, however legitimate they may be in philosophical contexts[11].

We have identified one confusion in the professional responsibilities of scientists that would be useful to dispel in order to reduce the harm that can by scientists who speak out of turn because they do not recognize that their responsibilities in the domain of practical decision are different from their responsibilities in the domain of scientific enquiry. The problem of scepticism beyond its proper bounds, which we have termed ‘Hyper-scepticism', is compounded by another failure of professional responsibility. This is also evidenced in the piece written by Warren (19, para. 2, l. .) He reports on one scientist, Stewart Franks from Newcastle University, who claims that he is ‘wary of mainstream claims projecting temperature changes over the next century' because we cannot explain natural variability in the climate ‘or predict when droughts will break'.

Franks assumes that models, which cannot make precise predictions of weather events, such as the break of a drought, cannot make at least probabilistic predictions of global trends, such as a tendency toward more frequent droughts in Australia[12]. This assumption is clearly false. The second law of thermodynamics states that entropy tends to increase in closed systems. There is ample evidence in support of this law, and every reason to expect that entropy in systems will tend to increase unless they take in energy from elsewhere, despite the inability of scientists to predict the precise course of the entropy of even closed systems.

Climate change models make abstractions. They focus on the ‘greenhouse effect', often excluding other factors that bear on the process. What they do consider is the balance between emission and absorption of gases in the atmosphere that affect the amount of radiation back into space from the Earth relative to inputs of energy from the Sun. From these models they predict trend increases, not the precise temperature at a given point on the Earth's surface at a given time. There is no reason to expect that predictions of overall trends in average temperature, or even predictions of increasing frequency of droughts, should be linked in any way with predictions of local weather at specific times.