I discuss two key issues for the analysis of early urban settlements: definitions, and comparative analysis. There is no ‘best’ definition of terms like city or urban. These are not empirical descriptions of the archaeological record; they are theoretical terms whose definition should match the research goals and questions of a study. Most archaeological definitions of city and urban use combinations of six dimensions of variability: size, functions, urban life/society, form, meaning, and growth. I then review seven reasons for archaeologists to pursue comparative analysis of past cities. Comparative analysis is necessary if we are to move beyond descriptions of individual cities to build an explanatory science of urbanism in the past.