The route from semantic to synthetic programming cannot be compared with a nice relaxed Sunday walk. The theoretical and methodological issues implied are of a complex nature, but the major obstacles seem to get pulled away within social learning across the scientific fields. In principle we consider that as a scientist we develop different sets of languages which need to be synthesized from time to time. We follow Hanappi (2008a) and compare the following language structures or ‘knowledge languages’ (compare Table 12.1) in order to synthesize the semantic programs already elaborated in novel ways. In this chapter we elaborate preparation work for future bottom-up models of evolutionary economic programs, where an exemplary agent-based simulation of institutional change is given in Chapter 13.