Abstract
The concept of similarity has occupied a key position in the interpretation of archaeological evidence since Thomson's Three Age System was formulated (Thomsen, 1836). Indeed, the writing of prehistory demands the use of this concept in relating evidence from stratigraphically distant horizons. Taylor, however, not only argued that too much emphasis could be placed on this comparative approach, but also claimed that it could be detrimental to the full recovery of archaeological information (W.W. Taylor, 1948). A similar dissatisfaction in Britain prompted Clark (1964, 1966, 1967) to adopt a 'conjunctive approach' (W.W. Taylor, 1948:7) exemplified in their work as economic prehistorians. This reappraisal however, has increased rather than diminished the need for procedured designed to relate assemblages in cultural terms. While the development of methods of analysis which assess the similarity between comparable items of different assemblages proceeds, it is stressed that sound theoretical principles, whereby the results of such analyses may be interpreted in the most plausible manner, must be adopted. Indeed it is urged that the common assumption that the degree of cultural similarity is directly proportional to the formal similarity, is by no means universally valid.
Considerable advances have recently been made in developing techniques to identify formal relationships by establishing the degree of 'proximity' between different assemblages of information. The ancillary problem of interpreting results in cultural terms has received relatively less attention.
This dissertation considers the application of methods of 'proximity' analysis to specific New Zealand assemblages, together with a discussion of the problems encountered in interpretation. The general implications of this research for prehistoric studies will also be considered. [Introduction]