Despite the findings of natural scientists and philosophers, the law of torts is still clinging on a strictly deterministic (in the Laplacian sense) idea of the world. Probabilistic considerations are not alien to the legal world, yet they are generally regarded as ad hoc exceptions to handle particularly complex cases. From this perspective, this thesis advocates the need for a theoretical shift. A probabilistic approach to reality should become the norm, whereas determinism should only be considered a heuristic tool when confronted with prima facie deterministic cases. In Chapter II it is shown that a strictly deterministic concept of causation is inadequate to face the intricacies characterizing modern litigation. In fact, the deterministic version of the ‘but for’ test necessarily creates frictions with the kind of evidence produced by modern science. The introduction of a purely probabilistic concept of causality is advocated and a distinction is drawn between the ex-ante and the ex-post probabilistic approach to causation. The former seems to be a better approach for lagged torts, whereas the latter is more appropriate to handle instant torts. Also, it is shown that in a probabilistic world the very concept of harm assumes a different meaning. Building on the new definition of harm introduced in Chapter II, in Chapter III it is suggested that the debate on the goals of tort law should be reconsidered. In a probabilistic world welfare maximization and corrective justice are not mutually exclusive, but must be regarded as necessary complements. In Chapter IV, it is argued that the problems created by a deterministic mindset stretch way beyond the analysis of causation. In fact, a Laplacian (deterministic) view of the world often prevents us from contemplating probabilistic solutions even when deterministic options have failed. From this perspective, Credit Rating Agencies (CRAs) are a perfect example. Both the legal and the economic literature have advanced solutions to improve CRAs incentives to issue accurate ratings. Yet, in most cases, the proposed solutions did not exploit the probabilistic nature of ratings, thus they were not framed exclusively in probabilistic terms. To the contrary, by designing a simple and legally workable strict liability rule it is possible to tie CRAs profits to the quality of their probabilistic predictions. In Chapter V, it is investigated whether the law and economics movement can be considered the answer to the indeterminacy of predictions haunting other sciences. Not surprisingly, the answer to this question is no. Even the extremely simplified world of tort law and economics is dominated by an indeterminacy principle.