For the problem that multi-granularity, hesitation, fuzziness exist in the linguistic information expressed by individuals, a method of group recommender systems with unknown attribute weights in a multi-granular hesitant fuzzy linguistic term environment is proposed. Firstly, the concept of multi-granular hesitant fuzzy linguistic term sets(MHFLTS) is defined. A variety of distance measures between two MHFLTSs are defined. Then, when the attribute weights are completely unknown, the objective programming model is established. The weights are obtained by using the Lagrange equation model. When attribute weights are incomplete unknown, the weights are obtained by solving a linear programming model. Finally, the movies recommendation is employed as an example to introduce the algorithm of the group recommendation, which shows the effectiveness of the proposed model in solving the group recommendation.