We consider a continuous version of the Hegselmann-Krause model of opinion dynamics. Interaction between agents either leads to a state of consensus, where agents converge to a single opinion as time evolves, or to a fragmented state with multiple opinions. In this work, we linearize the system about a uniform density solution and predict consensus or fragmentation based on properties of the resulting dispersion relation. This prediction is different depending on whether the initial agent distribution is uniform or nearly uniform. In the uniform case, we observe traveling fronts in the agent based model and make predictions for the speed and pattern selected by this front.