Stephan Hartmann (Tilburg) gives a talk at the Workshop on Mathematical Philosophy titled "Voting, Deliberation and Truth". Abstract: There are various ways to reach a group decision. One way is to simply vote and decide what the majority votes for. This procedure receives some epistemological support from the Condorcet Jury Theorem. Alternatively, the group members may prefer to deliberate and will eventually reach a decision that everybody endorses -- a consensus. While the latter procedure has the advantage that it makes everybody happy (as everybody endorses the consensus), it has the disadvantage that it is difficult to implement, especially for larger groups. What is more, a deliberation is easy to bias as those group members who make others change their mind may not necessarily be the best truth-trackers. But even if no such biases are present, the consensus may be far away from the truth. And so we ask: When is deliberation a better method to track the truth than simple majority voting? To address this question, we propose a Bayesian model of rational non-strategic deliberation and compare it to the straight forward voting procedure. The talk is based on joint work with Soroush Rafiee Rad.