The New York State Museum's Collections: A Case Study of the Present Sample
A logical beginning place for developing plans aimed at better managing collections is to determine the nature of the sample of the archeological record presently represented in curated collections. The nature of the present sample generally is unknown, largely because of problems in collections accessibility. A chronic problem in the management of archeological collections has been access to information about what is in a particular institution's collections. This situation is beginning to change as museums are being required to become more accountable for their activities, and as the information management capabilities of modern computers are applied to collections management. Support from programs such as the NSF Systematic Anthropological Collections Program is also a critical element of this change.
An NSF-sponsored collections inventory recently undertaken by the NYSM provided an opportunity to assess the nature of the curated sample of archeological material held by that institution (Sullivan 1987, 1989). The primary goal of the collections inventory was to make these collections more accessible for research, but the compiled data on collections composition also allowed recognition of information gaps.
The NYSM is a 150-year-old institution that has been "home" to a number of distinguished archeologists, including Arthur C. Parker, M. R. Harrington, and William A. Ritchie, who made collections that are still curated by the museum. It also serves as the repository for collections resulting from many cultural resource management related projects. The archeological collections include an estimated 1,000,000 objects and associated records with statewide coverage. As such, the NYSM archeological collections are the most extensive for the State of New York, and represent a century and a half of archeological research in the Northeast (Sullivan et al. 1990).
The representativeness of the State Museum's collections in relation to present knowledge of the nature of archeological resources in the State was assessed by asking the following questions. How well do the curated materials represent the archeological resources of the geographic region, i.e., the State of New York, from which the museum curates collections? Are all the known types of sites included? In what proportions? At what level of recording/investigation? How well are the investigations documented? Currently it is neither feasible to compile precise statistics nor to directly compare the contents of the NYSM collections with expected distributions of archeological resources in various regions of the State,5 but several patterns are suggested by the inventory data.
A series of maps and graphs showing collection distributions by geographic area, time period, and volume serve to illustrate the extent of coverage of the NYSM collections. First, as shown by Figure 1, a map of sites represented in the collections by county, the collections do not provide even geographical coverage across the State. Although potential distributional differences in site density in various areas of the State cannot be discounted, it is clear that the archeological resources from some counties and regions are very well represented and others are not represented at all.
The same kind of distributional pattern appears true for time periods, as shown in Figure 2. Although one would expect better representation for later prehistoric periods than for earlier ones because there are many more of the former, a graph of collections associated with general time periods indicates that certain periods are better represented than others.
Another collecting pattern is indicated when Figure 3, a map showing the volumetric distribution of objects by county, is compared with the distribution of sites by county shown in Figure 1. Differences in these distributions generally are related to the level of intensity of investigation. Those counties with few sites, but many objects, are indicative of large scale excavations having been conducted at one or more sites. In contrast, those counties with numerous sites, but few objects, are indicative of areas where some survey work was done, but little in the way of excavations. Thus the intensity of coverage also is variable across the State.
A fourth type of collecting bias was apparent in the course of the inventory work, but is difficult to illustrate quantitatively due to poor information about the nature of individual sites. This bias is related to kinds of sites that have been investigated intensively. By far the majority of sites that have been systematically and intensively investigated are large habitation sites or deeply stratified sites, the latter type usually being located in floodplain environments or rock shelters. Very little information exists in the NYSM collections from upland sites, lithic scatters, small habitation sites, and the like, even though such sites constitute the majority of the archeological record. There also is very little information from systematic regional survey. Instead, the collections are heavily biased toward those sites that are rich in terms of artifact content. This bias is largely a function of the history of the development of archeological research since stratified and artifact rich sites are necessary for building chronologies.
To summarize, the inventory of the State Museum's collections provides very basic qualitative and quantitative information about the representativeness of the present archeological database for New York State as contained in extant collections. Scrutiny of the areas of most comprehensive coverage suggests that because the bulk of the NYSM collections was made by its in-house archeologists, the collections are more reflective of these scholars' personal research interests rather than actual distributions of archeological resources. For example, Arthur C. Parker (1907, 1922) conducted much of his field research in extreme western New York and the Finger Lakes area. William Ritchie and Robert Funk continued work in the Finger Lakes region and also spent considerable time in the Hudson Valley and southeastern section of the State (Ritchie and Funk 1973; Funk 1976). This assessment is not intended to be a criticism of pursuit of individual research objectives by archeologists, but, as has been discussed, preservation of a database that can be used for future research is an important consideration for cultural resource management (Dunnell 1984; Glassow 1977; Milner 1987; Salwen 1981).
In general, cultural resource management projects have not begun to fill in the gaps in areal or site type coverage that are characteristic of the older collections. The lack of intensive investigations of what could be considered small-scale sites is especially noteworthy. This finding is consistent with a recent assessment of all archeological work conducted in the Upper Hudson region of New York State (Bender and Curtin 1990). Even though such sites are extremely important for modeling regional human landuse patterns, these sites often are difficult to interpret on an individual basis which makes case by case justifications for intensive investigation troublesome.
Other repositories may have collections that fill in some of the gaps observed in the collections curated by the NYSM. Data are not readily available to assess whether this may be the situation, although the New York Archaeological Council has made some progress toward a general inventory of extant collections for the State (Engelbrecht et al. 1988). Some of the gaps in areal coverage may correlate with the rate of modern development in certain regions. Areas with few archeological investigations may indicate areas with the least development activity. If this is indeed the case, then, in theory, the archeological record for these areas persists as intact sites.
The situation at the NYSM probably is typical in terms of the collections in its care vis-a-vis its area of coverage. The specific nature of biases in collections, however, may vary from institution to institution, and from region to region, depending upon the kinds of archeological problems historically investigated in specific areas and the local characteristics of archeological resources.