The sphere in which scientific hypotheses are debated may be termed the sphere of theoretical enquiry. As noted, dissent from prevailing opinion must be tolerated in this sphere. The consensus view today on global warming is that there will be significant increases in atmospheric temperatures by 2100 and that this will be due to increases in greenhouse gas emissions from human activities. Scientists who support alternative hypotheses, such as that the present trend of global warming may well be counteracted by changes in the albedo of the Earth or changes in solar radiation or, in any case, will only be due to causes other than human activity, are entitled to pursue their hypotheses and debate them in scientific forums, free of moral or other coercion.

However, this essential entitlement of scientific enquiry does not in itself support the complaint that dissent from the consensus view is now wrongly being swept aside in the sphere of public policy debate over global warming.  In forums of public policy debate, scientists offer opinions on what we ought to believe on scientific grounds for purposes of practical intervention. Debate over practical choices on the basis of scientific evidence belongs to what we term the sphere of practical decision. Our contention is that the professional responsibility of scientists in this sphere should not be confused with what is appropriate in the sphere of scientific enquiry. What is rationally subject to theoretical doubt or sceptical query is not necessarily rationally subject to practical doubt, or sceptical query of practical decisions. In fact, our contention is that what counts as scientifically responsible scepticism crucially differs in these spheres.

The fundamental distinction between scientific responsibilities in theoretical enquiry and practical decision may be illustrated by the role of expert advice in the context of a criminal trial. Prevailing scientific opinion is that HIV infection causes AIDS. By this scientists do not mean that HIV infection is a sufficient condition of AIDS. A few people with HIV infection do not go on to develop AIDS, even in the absence of intervention. Many people with HIV infection so far have not progressed to AIDS, after they started and maintained a burdensome daily regime of taking multiple anti-viral drugs from an early stage of their HIV infection. Notwithstanding these cases, which show that AIDS does not invariably follow HIV infection, it remains the case that infection with HIV causes AIDS. This amounts to the claim that there is some causal connection between HIV infection and the onset of AIDS which, under some description of each event, may be subsumed under some scientific law (Hunt, 2005: See also Davidson). In Mackie's ( ) terms, the onset of AIDS may be causally explained by citing the HIV infection as an in itself insufficient but necessary part of an unnecessary but sufficient condition for the onset of AIDS. In Cartwright's ( ) terms, HIV infection has a strong tendency to progress to AIDS. Having an immune system that suppresses the capacity of the HIV virus to multiply, invade and destroy the body's T-cells, or taking daily doses of anti-viral drugs, are merely countervailing factors to the expression of that tendency.

Given the prevailing scientific consensus, legislators ought to believe that HIV infection causes AIDS in each of the senses above. Given that AIDS causes death by making the body susceptible to pathogens and cancers, it is also reasonable for legislators to make it a criminal offence for a person to knowingly or recklessly infect another person with the HIV virus through unprotected sexual intercourse, or by other means such as sharing a needle for injecting heroin. Possible charges could range from a specific offence of knowingly infecting a person with a serious disease to a broader offence of endangering life.