In the literature on Non-Monotonic Logics, many examples of defeasible deductions infer that an arbitrarily chosen individual in a given set has the properties that we consider as “typical” for the elements of that set. In the paper we provide a logical framework, based on first-order logic, suitable for this kind of inferences. The main features of our approach are: 1) typicality of a given property for the individuals in a given set is represented by means of a binary determiner, 2) individual constants are evaluated on sets of individuals, so that a “range of uncertainty” is ascribed to them, 3) the adopted formalism and semantics allow formulas that adequately represent assertions like “it is highly plausible that the individual c has the property P”, and 4) the formal conditions under which “c has the property P” can consistently be inferred from the above assertion are determined, 5) non-monotonicity is achieved by changing the evaluation of the individual constants. In Appendices A, B and C, we provide a sound complete axiom system for the logic considered in the paper.