This volume

Chapter 2 presents an introduction to the physical landscape of the Taraco Peninsula and to the history of archaeological research in the immediate vicinity. The following chapter, Chapter 3, concerns the methodology employed in the collection of the settlement dataset. It includes a discussion of what I consider to be a significant methodological innovation in dealing with mixed surface collection assemblages. The method, termed frequency profile analysis, is a technique which allows for estimation of the relative proportions of various temporal components in a temporally mixed assemblage. It is used in this study in order to identify the sizes of various occupations of a given site, and thereby to permit calculations of rates of population growth or decline through time on both the site and regional scales.

The topic of Chapter 4 is the manner in which settlement data will be used in my analysis of Taraco Peninsula social evolution. The first half of the chapter consists of the development of a method of deriving a population index from the Taraco Peninsula settlement data. It should be emphasized that I make no attempt to estimate actual population values for settlements. Rather, I outline a method that produces population index values approximately proportional to population values. This means that my population index values may be used for internal comparisons, and perhaps for comparison between the Taraco Peninsula data and other Titicaca Basin datasets produced and normalized using the same techniques, but not for comparison with datasets from other parts of the world. However, rates of change within the Taraco Peninsula settlement system, especially rates of population growth, should be directly comparable cross-culturally. Also significant here is that my method has the property of reducing the disproportionate significance accorded to smaller and more ephemeral components of a settlement system by most population measures.

The most important part of Chapter 4 is my discussion of the meaning of intraregional variation in population growth rates. This discussion stems from the observation that at certain time periods certain sites or groups of sites seem to have grown at radically different rates, differences which cannot be attributed to differential fertility or mortality. The only possible explanation for these differences is therefore movement of population between sites and communities. This being the case, it becomes clear that the evidence of population growth rate differentials in effect provides a record of aggregate human decisions regarding the location of their residence. The disproportionate growth of one site, say, is evidence of a systematic bias in residential decisions on the part of the prehistoric inhabitants of the region. It therefore becomes possible to adopt an agent-centered approach to the interpretation of settlement data. The interpretive approach developed underpins the entirety of my subsequent historical reconstruction. The discussion begins on page . This section must be read in order to appreciate the arguments as a whole. I suggest that it be read before proceeding to the particulars of the archaeological sequence.

Following are Chapters 5 through 10, which present the Taraco Peninsula archaeological sequence itself. Chapters 5 and 6 discuss the Early and Middle Formative periods, when the settlement system was characterized by small, autonomous villages. Chapter 7 covers the Late Formative, and charts the local development of a multi-community polity and the early indications of Tiwanaku regional dominance. Chapter 8 describes the Tiwanaku period itself, when the Taraco Peninsula was reorganized as a heartland province of the Tiwanaku state, and Chapter 9 the period following its collapse. Finally, in Chapter 10, the conquest of the Titicaca Basin, first by the Inka empire and later by the Spanish, is discussed. Chapter 11 presents concluding remarks and observations.