Since the days in the 1920s when the New York State Museum’s Arthur Parker (1924) excavated at Coxsackie’s Flint Mine Hill, the flats, ridges, and stream-sides of eastern Greene County, New York, have drawn the attention of archaeologists. Archaeological sites are abundant here, as are important sources of material for prehistoric stone industry in the sought-after cherts of the Normanskill shale and Kalkberg limestone.
As he reworked his report on the West Athens Hill site (located in the Town of Athens), Bob Funk (2004:l-li) reflected on the draw that the stone sources exerted on early Native American pioneers, possibly including people of the Early Archaic period (8000-10,000 years before present; BP). He thus imagined a band of people discovering the chert source:
Having climbed the hill and commenced filling skin bags with the newly discovered Normanskill chert, a party of native men and women discarded two broken projectile points and made new ones to replace them (Bob had found two Early Archaic, bifurcated base point fragments on West Athens Hill). Before long, one of the women found old chert artifacts on the ground and called to the others. Bob further imagined the scene, the puzzlement and the emotions, writing:
“There was a brief flurry of excitement: had other beings, perhaps very different from the people of the band, visited the hill long before and made tools and weapons from its outcropping stone?”
West Athens Hill, best known as a Paleoindian site, was also visited intermittently during the Early Archaic and afterward. In nearby Coxsackie, Flint Mine Hill also saw Paleoindian visitors, while the fields around it abound with evidence of use during the Archaic period. While Early and Middle Archaic sites (8000-10,000 BP and 6000-8000 BP, respectively) have been considered rare in much of New York State, they are noticeably more frequent in eastern Greene County. This could be due to the attraction of the chert quarries, which may have centered the mobile settlement strategies of the Hudson valley’s hunting and gathering populations. This general idea— sometimes called the “lithic-centric” model— was developed by William Gardner (1977) to account for early settlement patterns in Virginia. It has been revived more recently in the context of studying the Early Archaic Hardaway site, North Carolina (Daniel 1998). Stated succinctly, this model views stone sources as significant reasons to establish seasonal settlements nearby. It was not enough to just visit the quarries; the quarries made these locales more attractive than others for settlement.
For Archaic hunters and gatherers, local food sources varied in type or abundance, and could be effectively exhausted, so movement to other sites was usually necessary after a period of time. However, the lithic-centric model sees mobility across the landscape as cycling back to important quarry locales on a regular, perhaps annual basis, even when substantial distances were involved. This settlement pattern would tend to concentrate artifacts such as stone projectile points in the quarry surroundings, while also dispersing these artifact types throughout larger regions, relatively far from the quarries (also, stone would be conserved carefully at distance from quarries, but used perhaps extravagantly near quarries).
This may be an important model for understanding the early periods of human settlement in Coxsackie and nearby towns such as Athens and New Baltimore, because when the larger region of eastern New York is considered, there appears to be a correlation between the abundance of the local chert resources, and relatively high frequencies of Early and Middle Archaic sites and projectile point finds.
Curtin Archaeological has performed several cultural resource surveys in this area, most of which have been supported by the Greene County Industrial Development Agency. One of the results of this work has been the identification of numerous Early and Middle Archaic period sites. Unlike many areas of New York State, the high frequency of local sites of this age makes it possible to go beyond noting the simple presence of population during these periods to making some inferences concerning ways of life.
One approximately 200-acre area has been studied most intensively through surface surveys and data recovery excavations. This area, like much of eastern Greene County, is a landscape in which ancient American Indians (going back to the Paleoindians) hunted, gathered, and likely procured materials for making more perishable technology (such as baskets, nets, and rope), performing many of the related tasks out in the woods and fields, away from habitual residences. They left broken, exhausted, lost, and stored stone artifacts in low densities across the landscape, and occasionally camped for shorter or longer periods, leaving small campsites marked by concentrations of stone artifacts as well as artifact manufacturing waste.
Over time, the study area captured abundant stone evidence of these activities and campsites. As part of a much larger area, it was visited by mobile people who usually dwelt elsewhere; people who moved in, and after a while moved out. During the earliest part of the Early Archaic period, about 10,000 radiocarbon years BP (and for several centuries thereafter), the visitors left isolated, Dalton-like and triangular projectile points, but may not have established campsites. By about 9,000 BP, a couple of small campsites appeared. These sites, associated with early side-notched and corner-notched projectile points, ranged in size from about 50 to 175 square meters. A little later, about 8500-8000 BP, people using bifurcated base points foraged in the area, but no associated campsites were found.
A different picture emerges in the Middle Archaic period, 6000-8000 BP. The projectile point technology shifted to points with stemmed hafts, such as Neville, Stark, Neville Variant and Merrimac points (cf. Dincauze 1976). While the larger landscape continued to be used for hunting and gathering, more campsites were established (possibly as many as 7, although 3 artifact concentrations are so close together they appear to represent one camp). In addition, the Middle Archaic sites varied more in size and content; they range from 75 to 1,000 square meters in size.
With respect to content, one Middle Archaic site has significant amounts of fire cracked rock in addition to chipped stone artifacts and manufacturing debris. Fire cracked rock forms when rocks shatter from exposure to heat, or heat and rapid cooling. Rocks which line hearths crack and shatter, as do rocks that were used in stone boiling. Stone boiling has been used by people who do not have fire-proof containers– such as ceramic or soapstone pots– for heating hot water. In stone boiling, food is prepared by raising the water temperature in containers such as skin sacks or wood or bark vessels by placing heated rocks in the water.
The variation in fire cracked rock occurrence may reflect seasonal differences in either resource processing or consumption patterns. For example, if nut oil was being rendered in heated water, a fall season of occupation would be indicated. If the concentration of fire cracked rock was associated with feasting, the seasonal aggregation of band segments may have occurred
In another interesting variation of site size and content, one of the Middle Archaic sites is composed of three adjacent artifact concentrations each in the size range of 100 to 225 square meters, suggesting that this site was occupied by three households or band segments (rather than just one), reflecting the formation of a community rather than a single, isolated campsite.
The variety in size and content among the Middle Archaic sites probably is associated with the use of the area by groups whose size and subsistence activity varied by season. The larger number and size of Middle Archaic sites, and the suggestion that varied seasons of occupation are represented indicate that population may have been growing, and the territories people used were becoming more localized.
One of the implications of these findings is that Early Archaic populations were more mobile (or mobile over greater distances) than Middle Archaic people. The inferred Early Archaic level of mobility would have dispersed Early Archaic populations over large territories, leading to fewer residential camps even in annually visited places such as the Coxsackie study area.
Seeming to support this idea, a study of the intensity of projectile point use may reflect differing levels of mobility. A minority of projectile points recovered in the study area have evidence of cutting, scraping or other uses in addition to the projectile function. A study of the variety and intensity of these uses (with low-power magnification) shows that Late Paleoindian and Early Archaic projectile points were used more intensively for these purposes than were Middle and Late Archaic period points. The hypothesis developed from this study is that compared to Middle Archaic points, Early Archaic projectile points had longer use lives, and were carried, maintained, and re-used for longer periods of time due to greater travel away from chert sources (for example, one of the Early Archaic points is made from western Onondaga chert from western New York).
The waste flakes from the Early and Middle Archaic campsites also may reflect long- or short-distance mobility differences. The Early Archaic campsites have higher frequencies of large flakes than do the Middle Archaic sites. Since larger flakes are generated in earlier stages of stone tool manufacture, this difference probably means that Early Archaic populations moved on to other sites after roughing out quarried stone material, while Middle Archaic populations more often refined and even finished stone tools within the study area. Moreover, some of the Middle Archaic campsites have 3 to 6 times more chert waste flakes than the Early Archaic sites (this reflects an aspect of the lithic manufacturing process: as waste flakes become smaller in later stages, they also are generated in greater numbers). At the same time, some of the Middle Archaic sites seem not much different than Early Archaic sites with regard to waste flake frequency per square meter, reflecting that while Middle Archaic populations may have been more localized, they were still frequently mobile, and sometimes moved on while substantially staging stone tool manufacture.
The Coxsackie study area is only a small landscape in a much larger region, but it provides a case of a hypothetical pattern of change from Early Archaic to Middle Archaic times. If this area is typical, then Early Archaic populations journeyed far and used stone tools rather exhaustively while away from the Coxsackie sources of stone; but returned regularly at least for short periods, hunting, gathering, and replenishing stone technology while living in small camps. By the Middle Archaic, people were establishing more and sometimes larger sites in the area, possibly because population was larger, and people sometimes lived in larger groups. Different Middle Archaic subsistence pursuits may have been involved, and long-distance, seasonal relocation seems to have been reduced since the Early Archaic.
If reflective of the lithic-centric model, the Early and Middle Archaic patterns appear to represent two different levels of human association with stone sources. While the Early Archaic pattern seems like the lithic-centric model as often discussed in the literature, the Middle Archaic pattern may be associated with differential access to stone resources due to reduced long-distance mobility, and the centering of some settlement patterns in areas lacking chert quarries. More data from other locales and regions are needed to examine whether these perceived differences are real.
If real, the pattern of change may be associated with a warming post-Ice Age climate; and after about 8000 BP, more abundant food resources in a forest increasingly populated by hardwood trees, many of which produce nuts and acorns, a.k.a. mast. Increased mast production provided more food for prey species such as deer, and eventually for direct human consumption. Oak and other nut-producing trees were becoming significantly more abundant by the appearance of the Middle Archaic.
Although this may be imagined as a rather simple cause-and-effect relationship— a pattern of people prospering in good times— it may have been more complicated. For example, sea-level rise was another effect of the warming climate that promoted the Northeastern increase in nut-bearing trees. We can only speculate about whether, over the long run, chain reactions of altered population density and patterns of social interaction resulted from human migration inland as the Atlantic (and its bays and estuaries) transgressed the shore-lines.
Daniel, I. Randolph, Jr.
1998 Hardaway Revisited: Early Archaic Settlement in the Southeast. University of Alabama Press, Tuscaloosa.
Dincauze, Dena F.
1976 The Neville Site: 8000 Years at Amoskeag. Peabody Museum of Archaeology and Ethnology Monographs No. 4. Harvard University, Cambridge, Massachusetts.
Funk, Robert E.
2004 An Ice Age Quarry-Workshop: The West Athens Hill Site Revisited. New York State Museum Bulletin 504, Albany.
Gardner, William M.
1977 Flint Run Paleoindian Complex and Its Implications for Eastern North American Prehistory. In Amerinds and Their Paleoenvironments in Northeastern North America, edited by Walter S. Newman and Bert Salwen, pp. 257-263. Annals of the New York Academy of Sciences, Volume 288, New York.
Parker, Arthur C.
1924 The Great Algonkin Flint Mines at Coxsackie, New York. Researches and Transactions of the New York State Archeological Association 4(4).