For some time there has been increasing interest in the problem of monitoring the occurrence of topics in a stream of events, such as a stream of news articles. This has led to dierent models of bursts in these streams, i.e., periods of elevated occurrence of events. Today there are several burst definitions and detection algorithms, and their dierences can produce very dierent results in topic streams. These definitions also share a fundamental problem: they define bursts in terms of an arrival rate. This approach is limiting; other stream dimensions can matter. We reconsider the idea of bursts from the standpoint of a simple kind of physics. Instead of focusing on arrival rates, we reconstruct bursts as a dynamic phenomenon, using kinetics concepts from physics -- mass and velocity -- and derive momentum, acceleration, and force from these. We refer to the result as topic dynamics, permitting a hierarchical, expressive model of bursts as intervals of increasing momentum. As a sample application, we present a topic dynamics model for the large PubMed/MEDLINE database of biomedical publications, using the MeSH (Medical Subject Heading) topic hierarchy. We show our model is able to detect bursts for MeSH terms accurately as well as efficiently.