In a 2003 IEEE-FUZZ Conference paper, Mendel proved that to use a type-1 fuzzy set model for a word is scientifically incorrect. Here he explains that to use an interval type-2 fuzz set model for a word is correct, and provides a first-order uncertainty model for a word. The paper describes two methods for solving the following inverse problem: Given data that are collected from a group of individuals about a word (on a scale of 0-10), map the data and its associated uncertainties into the footprint of uncertainty of an interval type-2 fuzzy set model for the word.