This paper presents a multiobjective Lamarckian memetic approach to fuzzy job-shop scheduling. In particular, a specialized local search based on the N2 neighborhood structure is used to improve triangular fuzzy makespans across their three solution graphs in parallel. On the other hand, the NSGA-II is used to further minimize the three defining points of the fuzzy makespans. An extensive experiment was conducted to confirm the superiority of the algorithm compared to both the single-objective memetic and multiobjective genetic methods. A new analysis scheme is also introduced, which shows that regardless of the expert's attitude toward uncertainty quantification, the proposed approach would consistently find statistically better schedules in terms of expected makespan.