Chance as a new foundation for logic

With this paper I start the investigation of the following claim: logic (and mathematics) is reliable because it is rooted on the laws of chance. To show this I introduce a stochastic semantics for logic where propositions are interpreted into sets of observables (i.e. sets of trajectories of suitable stochastic processes) whose elements describe some symmetries of chance. A validity theorem is proved for this semantic.