This paper describes the role of Modeling and Simulation (M&S) as a critical tool which must be necessarily used for the development, acquisition and testing of autonomous systems. To be used effectively key aspects of development, acquisition and testing must adapt and change to derive the maximum benefit from M&S. We describe how development, acquisition and testing should leverage and use M&S. We furthermore introduce and explain the idea of testable autonomy and conclude with a discussion of the qualities and requirements that M&S needs to have to effectively function in the role that we envision.