In this paper, a distributed control law is proposed for multirobot haptic teleoperation, based on consensus algorithms and acceleration control. This novel technique, called haptic consensus, enables robots on a network to agree on their accelerations, velocities and positions as well as a common force feedforward, by simply communicating the state data and force measurements of the robots on the information network. The key technology developed in this paper to realize haptic consensus is the acceleration consensus algorithm and a force consensus filter, which computes an average of the force measurements of the robots in a distributed way. Using these tools together we obtain the haptic consensus algorithm which enables multilateral teleoperation. In this paper, we also show experiment results for haptic consensus on a teleoperation network, with multiple telerobots, enabling multiple operators to agree on a common teleoperation task by tactile sensation.