Abstract: The two-tuple fuzzy linguistic representation model is a new approach to linguistic information aggregating, which make the aggregating result more precise and the language operation process easier and can avoid losing or distorting some of the original preference information in the process of aggregating linguistic information. The consistency of two-tuple linguistic judgement matrix is studied. Describing the definition of two-tuple, the rules to compare two different two-tuples are given. Several judgement approaches to the additivity consistency and satisfaction consistency of the two-tuple linguistic judgement matrix are introduced according to their concepts. Furthermore, the relationship between additivity and satisfaction consistencies is analyzed. An example is given to illustrate the use of the proposed approaches.