Typological analysis is a fundamental process in archaeological research, and is defined as the systematic arrangement of material culture into types based on similarities of form, construction, decoration or style, content, use, or some combination of these. Before the advent of absolute dating techniques, typological analysis, in conjunction with stratigraphic excavation, was the only means by which archaeologists could develop cultural-historical sequences or otherwise measure the passage of time. The assumptions of this approach are that, within a region, artifacts of similar form or style are near to one another in time, and that stylistic change is likely to be gradual or evolutionary. Early pioneers of typological analysis included the Swedish scholar Oscar Montelius, who in the late nineteenth century refined the chronological sequence of the Scandinavian Neolithic and Bronze Age; the English archaeologist Sir Flinders Petrie, who in the late nineteenth century developed seriation; and the American archaeologists Nels Nelson and A. V. Kidder, who in the early twentieth century developed a series of regional chronologies in the American Southwest based on stratigraphic excavation combined with careful description of the stylistic variation of ceramics and their arrangement into types.
A “type” is a set of characteristics (known as attributes by archaeologists) of an object that are consistently associated within one another. An “attribute” is a characteristic of an artifact that cannot be divided into smaller constituent units. Examples include raw material, color, size, and so forth. Attribute states are mutually exclusive instances of attributes. Pottery vessels, for example, can be sorted into use categories such as bowls, jars, plates, and other forms, each of which is an attribute state.
Types based on attributes related to use are known as functional types. Stone tools, for instance, can be placed into functional categories using attributes such as the angle of the working edge, and patterns of damage found along it. Another common approach to typology is to create descriptive or morphological types, wherein the goal of classification is to create an exhaustive description of artifacts so that they may be better compared with materials from other sites. Finally, types are frequently used to place objects in time, and thus are diagnostic of particular time periods.
Typologies are arbitrary constructs created by archaeologists to address specific questions that arise from the research process. Because typologies are created from such questions, no single classification of an artifact places it into its “correct” or “best” type. As research interests and priorities change, so may the content of types. Thus, objects may be found in multiple types depending on the needs of the researchers to address different questions. Although typological analysis is historically associated with chronology building and the concept of style, it is not limited to these notions.
As traditionally practiced, typological analysis is often viewed as more an art than science. Long familiarity with the materials under study and an intuitive “knack” for distinguishing between relatively small differences in attribute states are often seen as prerequisites for a successful analysis. To some, however, this intuitive approach to typology lacked scientific rigor and, worse, was seen as highly arbitrary in the choice of attributes. In response, a number of mathematical, statistical, and formal techniques have been applied to typological analysis. Albert Spaulding in 1953 published an influential paper that introduced archaeology to the chi-square statistic. Since then, other techniques to create types have been used, including cluster analysis, ordination methods, factor analysis, and contingency table analysis. Although in many instances these methods have proved to be of value, there nevertheless continues to be significant debate over the degree to which they are truly useful to or even necessary for typological analysis.
This debate, however, is much deeper than simply an argument about the relative merits of different techniques. It is also an appraisal of the meaning and goals of typological analysis in archaeological research. In his 1953 paper, Spaulding made a strong claim that artifact types were inherent in material culture and could thus be discovered through a statistical analysis of their mutual association. Furthermore, he argued that nominal scale attributes (color, shape, etc.) were more appropriate for analysis than quantitative attributes (size, weight, etc.) since they were likely to be a direct reflection of the ideals and values of those who made the artifacts. The goal of typological analysis, then, was to discover these ancient folk classifications (labeled by other archaeologists as “mental templates”). Types discovered using this approach to classification are seen as fundamental and immutable.
Reaction to these assertions has centered upon the validity of two assumptions: the logical priority of nominal scale attributes and the reality of types generated by their use. Many have argued that, although nominal attributes are often useful, they are not necessarily fundamental to typological analysis, and furthermore, their use does not permit the archaeologist to effectively explore the variability in a collection of artifacts. Vessel size might be of significance to a typology, but the use of nominal attributes does not provide a useful means by which variation in size can be easily defined or observed. As to the reality of types, though it is possible that nominal-scale attributes may well capture aspects of the way ancient peoples perceived some artifact, it is not necessarily the case that this was so, and from a scientific perspective, there is no way that such an assertion could ever be verified.
No matter how types are constructed, what is of greatest importance to the archaeologist is how they are interpreted and used. Therefore, in typological analysis, it remains the case that the archaeologist must think carefully about the questions being asked, and on that basis select the attributes and attribute states most likely to further that end.
Adams, William Y., and Ernest W. Adams. Archaeological Typology and Practical Reality, 1991. Cowgill, George L. “Artifact Classification and Archaeological Purposes.” In Mathematics and Information Science in Archaeology: A Flexible Framework, edited by Albertus Voorips, pp. 61–98, 1990. Dunnell, Robert C. “Methodological Issues in Americanist Artifact Classification.” In Advances in Archaeological Method and Theory, edited by Michael B. Schiffer, pp. 149–207, 1986. Dunnell, Robert C. Systematics in Prehistory, 1971. Spaulding, Albert C. “Statistical Techniques for the Discovery of Artifact Types.” American Antiquity 18 (1953): 305–314. Whallon, Robert, and James A. Brown, eds. Essays on Archaeological Typology, 1982.