This paper emphasises the importance of abduction in many enquiries in which the validity of norms or general principles are gathered from facts. First, it is shown how the kernel of abduction is an argument of the form: if a given hy-pothesis is true, then some consequences take place; in fact the consequences take place; therefore, probably, the hypothesis is true. Here the consequences play the role of hints or signs of the hypothesis. Moreover, abduction strength is analysed in the light of Bayes’ famous theorem. Next, the author shows how abduction may be employed in enquiries both of sociology of law and of legal science: in the first case in the form if a norm is valid, then some behaviours hold; in fact they hold; thus, probably, the norm is valid; in the second case in the form if a legal system is inspired by a given principle, then some statutory laws must hold; actually they hold; thus, probably, the legal system is inspired by the principle. Some remarks about further uses of abduction within the frame of law as well as in other fields of investigation conclude the paper.