Influential members of the urban planning profession have developed certain ideas about new town design, including notions such as self-containment, social balance, and the neighborhood unit. These parallel, to some extent, concepts that have emerged from the field of community sociology. Efforts to put these ideas into practice have fallen far short of the mark. Without more sophisticated implementation mechanisms, better theories of social interaction at the neighborhood level, and new approaches to citizen participation, efforts to build new towns are likely to remain severely crippled. The aim of this paper is to summarize past efforts to translate implicit theories of social organization into actual new town designs. The possibilities of closing the gap between theory and practice through the use of more explicit forms of social experimentation are discussed in the context of the fledgling new towns program in the United States.