Adrian's point taken. But can you outline for me how le Chatilier's principle would work in this case, and what the system actually comprises? I can think of two systems which don't behave in this way right off: a marble on top of a hill and a collapsing star.
In chaos theory, one can find all sorts of systems which change their behaviour radically given a small perturbation....
A major problem here is that only two of a multiplicity of factors have been considered. In isolation things would go as suggested with runaway positive feedback but, for imstance, extra water vapour would condense at height, forming clouds which increase earth's albedo and so reduce the amount of heat absorbed from the sun. Clouds also lead to increased precipitation which will a) absorb CO2 to form carbonic acid which will increase weathering of rocks and hence tie up carbon in new sediments and, b) fall as snow in high latitudes, increasing the albedo even further.
Add in all the other factors, known, unknown and Rumsfeld, which are "parameterised" (i.e. guessed) in climate computer models and the possible combinations/permutations are well beyond any reasonable predictive model.
Of course, not all the feedbacks will be negative - there is always a faint chance that the True Believers are right, but some 4.5 milliard years of experience suggests they are not!