An adaptive fuzzy sliding control (AFSC) scheme is proposed to control a passive robotic manipulator. The motivation for the design of the adaptive fuzzy sliding controller is to eliminate the chattering and the requirement of pre-knowledge on the bounds of the errors associated with the conventional sliding control. The stability and convergence of the adaptive fuzzy sliding controller are proven both theoretically and practically by simulations. A three-link passive manipulator model with two unactuated joints is derived to be used in the simulations. Simulation results demonstrate that the proposed system is robust against structured and unstructured uncertainties.