This paper introduces communication systems (CS) as a unified model for socially intelligent systems. This model derived from sociological systems theory, combines the empirical analysis of communication in a social system with logical processing of social information to provide a general framework for computational components that exploit communication processes in multiagent systems. We present an elaborate formal model of CS that is based on expectation networks and their processing. To illustrate how the CS layer can be integrated with agent-level expectation-based methods, we discuss the conversion between CS and interaction frames in the InFFrA architecture. A number of CS-based applications that we envision suggest that this model has the potential to add a new perspective to Socionics and to multiagent systems research in general.