Confirmation in Archaeology

Strong claims for the hypothetico-deductive method of confirmation have been made in recent works by new archaeologists. Although the explicit formulation of hypotheses and the testing of these hypotheses by observation and experiment are important to scientific archaeology, emphasis on deductive methods is misplaced. This chapter presents an alternative account of confirmation along Bayesian lines. This account recognizes the importance of plausibility arguments for assigning probabilities prior to testing. This approach to confirmation also provides a framework for distinguishing between tests that significantly alter the status of a hypothesis and those that make little difference. Popperian claims about the importance of falsification can be accommodated within this pattern of confirmation, and it has the distinct advantage of recognizing the role of alternative hypotheses. The chapter discusses the problem of the limits of analogy as a method of obtaining knowledge of the lives of archaeologically known peoples.