Beyond mediocrity: how common is life?

The probability that life spontaneously emerges in a suitable environment (abiogenesis) is one of the major unknowns in astrobiology. Assessing its value is impeded by the lack of an accepted theory for the origin of life, and is further complicated by the existence of selection biases. Appealing uncritically to some version of the 'Principle of Mediocrity' - namely, the supposed typicality of what transpired on Earth - is problematic on empirical or logical grounds. In this paper, we adopt a Bayesian statistical approach to put the inference of lower bounds for the probability of abiogenesis on a rigorous footing, based on current and future evidence. We demonstrate that the single datum that life has appeared at least once on Earth merely sets weak constraints on the minimal probability of abiogenesis. In fact, the a priori probability assigned to this event (viz., optimistic, pessimistic, or agnostic prior) exerts the strongest influence on the final result. We also show that the existence of a large number of habitable worlds does not necessarily imply, by itself, a high probability that life should be common in the universe. Instead, as delineated before, the choice of prior, which is subject to uncertainty (i.e. admits multiple scenarios), strongly influences the likelihood of life being common. If habitable worlds are uncommon, for an agnostic prior, a deterministic scenario for the origin of life might be favoured over one where abiogenesis is a fluke event.