Models
Citation: Winterhalder, B.P. (2002) Models. In Darwin and Archaeology: A Handbook of Key Concepts, Chapter 12 (RSS)
Internet Archive Scholar (search for fulltext): Models
Tagged:
Summary
According to Winterhalder, models is representing observed or hypothesized relationships of structure and function in simplified or abstract from and they can change a complex system or process to make it more accessible one. The models are ubiquitous in archaeology, biology, and related historical evolutionary sciences. History of using models has three steps. At first, use of explicit, quantitative models in ecology, e.g. simulation model, can be dated to the last 50 years. The second watershed occurred by ecologists challenging the reigning descriptive naturalism, evolutionary ecology. This evolutionary track used hypothetico-deductive methods and problem-oriented fieldwork. In other words, the model adopted the premise of natural selection. And then archaeologists and anthropologist started to use the similar tracks of (cultural) ecologists. Even those two kinds of social scientists use the approach more. As third development, the archaeologist and other social scientists adopt agent-based modeling.
There are several kinds of models. They can be divided into two main categories, scale models and heuristic models and the heuristic models are also divided into several sub-models, theoretical, analogic, and empirical models. Scale model can be represented by a topographic map or architectural miniature and it can show perceptual illusion of realistic scales. Heuristic models are problem-solving mechanism. Among them, theoretical models provide an operational bridge between abstract beliefs and concrete magnifications, analogic model derive the power of problem-solving from comparison with phenomena. Empirical model comes from the referent phenomenon itself.
The models have some advantages. They can be used at all stage of scientific explanation. They are good for defining and isolating a problem, organizing thought, advancing understanding of data or direct attention to relevant data yet to be gathered, communicating ideas, devising hypotheses and tests and making prediction. The models can also be divided into four categories based on the state of knowledge in a subject, (1) empirical: numeric, econometric, statistical; inductive, (2) exploratory, analogic, (3) theoretical deductive, (4) mature, high confidence models. Practically, one model can be applied into three different modes: the evaluative, interpretive, and normative.
To assess the empirical fit of hypotheses with observations, four classes can be suggested: (1) the model should strictly faithful to its antecedent theory, (2) the model should display internal logical consistency, (3) empirical evaluation of a model may entail direct testing of its assumptions, and (4) model should be a direct comparison of predictions with observations.
In addition, the author discusses about several philosophical issues such as difference between models and metaphors, narratives and reductionism. And he introduces three archaeological cases of modeling relating prehistoric period such as residential and field processing, production risk and social integration, and Paleolithic population ecology.
