7. Conclusion

In a necessarily schematic way, we have presented and contrasted here two extreme models of risk analysis: the so-called "standard" model, based essentially on experts and the principle of prevention of proven risks, and the so-called "constructivist" model, involving civil society and the application of a precautionary principle proportionate to plausible risks. We would like to make four final comments.

– First of all, with regard to the objective stated in the introduction, it is clear that any comparisons between models must be made with reference to the achievement of this overall objective, and not phase by phase. There's no point in having the best risk assessment system, or the best risk management, or even the best risk communication, if a weak link, or a poor link between these links, handicaps the final result.

– In relation to...