In this chapter I do not intend to plow new ground, but take a retrospective stance through explicating a particular methodology which has been applied to the classification of pottery in the South American tropical lowlands. I shall refer to this methodology as structural classification, since it is modeled after the methodology of descriptive linguistics. It is not to be confused with structuralism or structuralist analysis (Leone 1982) as those terms are currently used in a Lévi-Straussian sense in the archaeological literature. It is more akin to what has been called componential analysis, or ethnoscience in anthropological literature of the past four decades. Donald Lathrap was a strong advocate of this methodology, and his unpublished thesis (Lathrap 1962) is the earliest example of a thorough application of this methodology to an archaeological data set.

I write this with a slight feeling of apprehension, knowing that classification, particularly as it applies to ceramics, has historically been a hotly debated subject among archaeologists. As Spaulding (1982: 1) astutely observed, “Basic concepts and their implications are often controversial simply because they are basic.” If I touch some raw nerves, it is not by intention. My objective is to elucidate a particular methodology which has been used and developed by archaeologists working in the South American tropics over the past three decades.