The main objective of this paper is to examine the interpretation of so-called "equative tautologies" in adults and children. An experiment designed to assess the capacity of adults and children ages 7-9 to calculate implicature from tautologies is discussed. The prediction of the "Radical Semantic" account of tautologies (Wierzbicka 1987, Gibbs & McCarrell 1990) that human referent tautologies (e.g. "a plumber is a plumber") are easier to interpret than concrete referent tautologies (e.g. "a snack is a snack") was not confirmed by data from 23 adult subjects. Furthermore, the child data suggest that children do not interpret tautologies in an adult-like way. When presented with tautological statements, 7-year-olds tended to rely on their own preferences and knowledge of others' preferences rather than computing a conversational implicature as adults do, but this tendency decreased with age. Several explanations are provided to explain the poor performance of children compared to adults in the experiment, with suggestions for future work on the comprehension of tautologies.