Sneak Peek: Lithic Analysis Research for the NYSAA 2015 Annual Meeting

    The 99th Annual Meeting of the New York State Archaeological Association begins the evening of Friday, May 1, 2015 and runs through Sunday morning May 3.  The meeting, which includes the annual conference on Saturday and Sunday, will be held at the Ramada Inn, Watertown, New York.  Kerry Nelson, Meadow Coldon and I will be presenting our paper on Saturday morning at 9:50.  Here, Fieldnotes gives you a preview plus a small bonus:  a little additional information and “big picture” analysis that is not included in the conference paper due to time limitations.

Meadow and Ed sorting and analyzing bifaces from Fernlea sites 1 and 2.

Our paper is titled “Problem Orientation and Approach to Lithic Analysis:  Examples from New York State.”  This paper is an opportunity to briefly summarize some ways we have approached studying chipped stone tools from the A. Shafer site in Cobleskill, the Fernlea 1 and 2 sites in Coxsackie, and the Esmond 2 site in the Town of Malta, just outside of Saratoga Springs.  

No evidence was recovered to allow an age estimate of the A. Shafer site, but enough information was recovered to determine that this site forms part of a pattern of local chert exploitation that seems to characterize Cobleskill and adjoining Richmondville.  Recurrent similar finds in this area indicates that Native Americans in ancient days obtained useable chert for stone tools by shattering cobbles recovered from local streams, such as Cobleskill Creek.  Although the full time-range associated with this pattern probably has not been identified (and the A. Shafer site may pre- or post-date this range), time sensitive artifact types associated with this mode of obtaining chert in the Cobleskill area range in age from the Early Archaic period to the very end of the Archaic period (the latter often called the Transitional period).  The estimated age range is from about 9,000 to 3,000 radiocarbon years ago.  

At the A. Shafer site (and several others in the Cobleskill area), the shattered, stream-derived chert often is of a variety called “Esopus”.  Esopus chert was eroded from the local Devonian age bedrock before deposition in nearby streams.  Other chert, most often derived originally from Onondaga limestone, was obtained from glacial till before being brought to the A. Shafer site.  Most often, at A. Shafer and other local sites, the stone tools were technologically simple types such as utilized flakes and cores; or were manufactured with small measures such as isolating sharp graver spurs or retouching flake edges to make scrapers.  The recurrence of a pattern involving the use of local stream chert sources and the manufacture of expedient tools on the spot suggests that the Cobleskill vicinity may have been a sort of district where chert raw material was widely available and could be obtained predictably for day-to-day activities, or to procure and process locally available resources.  Lest readers imagine small, nuclear families of hunter-gatherers wandering occasionally into this district, I should point out that our paper takes the position that Cobleskill in prehistory probably was a repeatedly-used target destination for obtaining resources valuable for food or to construct technology, possibly visited repeatedly for work or short-term camping by multi-generational groups led by women (cf. Versaggi et al 2001).  

Gravers from the A. Shafer site.

In contrast to the A. Shafer site, the Fernlea 1 and 2 sites, as well as the Esmond 2 site feature biface technology, a more complicated technology requiring a staged process brought into material existence through a relatively complex inner-vision that involves significant foresight and motor skills.  Biface manufacture features the reduction of larger objects into smaller through the execution of controlled fracturing designed to make the biface thinner while reducing its width as little as possible.  As anyone who has tried to make an “arrowhead” knows, it is an understatement to say that this requires skill and experience.  Finished bifaces optimize the stone tool production process for transportability, and often provide “preformed” objects that, with a little more work, can be transformed into finished tools such as projectile points and stone knives.  Meadowood cache blades, such as those found at the Esmond 2 site, and which were manufactured across much of the breadth of upstate New York, are some of the most sophisticated late stage bifaces that occur anywhere.  They simply require notching to make them into projectile points, or a little pressure flaking of the base to turn them into endscrapers.  

The Fernlea 1 and 2 sites date to the Early Archaic period about 8,000 to 9,000 radiocarbon years ago, while the Esmond 2 site dates to the Early Woodland period about 2,000-3,000 years ago.  The Fernlea sites occur close to stone sources, while the sources of chert for Esmond 2 are several to dozens of miles away.  Although these archaeological sites are not contemporary with each other, their positioning in relation to stone sources provide an interesting contrast that helps to understand aspects of biface production and use.  We will return to this presently as this discussion moves beyond what is covered in the NYSAA paper.  First, however, some more information about the Esmond 2 stone artifacts is in order.

As a preliminary study, we have examined a sample of 65 chipped stone artifacts from the Esmond 2 site.  Fifty-five of these are made of Onondaga chert (with sources at least 30 miles away), and most of these are recognizable types associated with Meadowood technology.  The 10 non-Onondaga chert artifacts tend to be projectile point types made of local chert; in age, they span the same or a slightly longer time frame as the Meadowood phase of the Early Woodland period.  We did not know when we began our analysis that all of the Meadowood artifacts would be made of Onondaga chert.  Moreover, we did not know whether early stage bifaces were transported over a long distance to the Esmond 2 site and worked there into Meadowood-type tools such as points and scrapers, or whether late stage bifaces were brought there for the small modifications needed to make these tools.  We found that only about 3 each of Stage 1 and 2 bifaces (the earliest in the production sequence) occurred, while there were 12 Stage 3 bifaces and 18 Meadowood cache blades (Stage 4 bifaces).  This indicates that either it was mainly late stage Onondaga chert bifaces imported into the Esmond 2 site, or there was a high rate of success in transforming early stage bifaces to late stage bifaces:  that is, the observed pattern could mean that bifaces rarely break in the early stages, so that early stage bifaces are rarely found in comparison to late stage biface frequency.

3 stages of biface reduction out for analysis. Fernlea sites 1 and 2.

This is where we return to the information available from the Fernlea 1 and 2 sites, and in doing so, move into a comparison beyond what we cover in the NYSAA paper.  At the Fernlea sites, we recovered a sample of 42 bifaces that were large and complete enough to classify by reduction stage.  The Stage 1 to Stage 2 to Stage 3 biface ratio was 12:17:13 (there are no Stage 4 bifaces in this Early Archaic period technology).  Since the Fernlea sites are near the chert sources, this ratio is a strong indication that bifaces break in every stage of manufacture, and quite frequently in the early stages at the locus of production.  Even though many bifaces (and perhaps especially late stage bifaces) were removed when people left these sites, it seems a fair assumption that manufacturing into the late stages occurred near chert sources, and the populations of sites such as Esmond 2 that are located far from chert sources acquired late stage bifaces to the near-exclusion of others (there is, admittedly, a uniformitarian assumption here about biface technology, spanning thousands of years, but this seems reasonable since the manufacturing process continued to have similar goals, and is governed by the physics of fracturing chert).  

The Stage 1 through Stage 4 Onondaga chert biface ratios at the Esmond 2 site, 3:3:12:18 indicates that bifaces typically arrived as Stage 3 or Stage 4 bifaces, and that some of the significant stone-work at the Esmond 2 site involved transforming Stage 3 bifaces into Meadowood cache blades (Stage 4 bifaces); as well as Meadowood cache blades into other objects (such as endscrapers) as needed.  Very few Meadowood type projectile points or point fragments have been found at Esmond 2 or the adjoining Esmond 1 site, which suggests that at Esmond 2 it was more important to have Meadowood cache blades than projectile points (Although Meadowood Stage 4 bifaces are called cache blades, they are often found in other contexts, such as living areas; at Esmond 2 they appear to be dispersed across the site of a settlement rather than clustered in caches). 

Now, to add some interpretation not covered in the NYSAA paper:  Archaeologists sometimes entertain the view that Meadowood cache blades may be more than a step in making bifacial stone tools.  The distinctive, Onondaga chert blades may be something of more abstract symbolic value (I hesitate to say an early kind of commodity, because this understates their likely importance as gifts of political significance).  In a stripped-down version of this idea, Meadowood cache blades may have been the main form in which Onondaga chert was acquired, exchanged, or presented as gifts as aspects of forming alliances between important regional leaders or communities (I recommend Tache 2011 to explore the dimensions of Meadowood interaction further).  The Esmond 2 site may represent a location important in the process of distributing Onondaga chert beyond its nearest source areas in the Schoharie valley or other locations south of the Mohawk River.  The Esmond 2 population may have taken on this special role in relation to neighboring populations located farther from the Onondaga chert sources.  

Meadowood phase sites are somewhat rare in the upper Hudson region.  I tend to think of the Esmond 2 site not as a site evincing a Meadowood phase migration but rather as a local, Early Woodland community that participated in Meadowood technology as part of a larger network of interacting communities that managed their mutual inter-relationships at least in part with respect to the exchange of exotic, Onondaga chert items such as Meadowood cache blades.  This sense is based upon the occurrence of the Meadowood phase lithics at a site (Esmond 2) that seems to rely on projectile point types of local (Normanskill) chert made in arguably contemporaneous, non-Meadowood forms.  So there seems to be a hint of the local and traditional maintained alongside an abundance of the new and exotic.  

Meadowood cache blades.

Elsewhere, caches of Meadowood blades occur south and east of the upper Hudson region in places where otherwise there is, at best, a weak Meadowood presence (for example, in the Delaware valley where caches have been found at mortuary sites; Tache 2011:53).  In non-cache contexts, beyond a far distance from demonstrable Meadowood technology-makers, Meadowood cache blades (due to Onondaga chert scarcity) may have often ended their use-lives transformed into tools without regular replenishment.  For example, at Early Woodland settlements such as those found in the Delaware valley by Kinsey (1972), Kraft (1975), and Werner (1972), Meadowood assemblages are small and dominated by Meadowood-type projectile points made of imported Onondaga chert, while Meadowood cache blades are scarce.  This type of assemblage is the opposite of the Esmond 2 site, where Meadowood cache blades are abundant, and Meadowood points scarce.  In my opinion (and here I am far indeed from the NYSAA paper), the Delaware valley Meadowood artifact assemblages do not indicate Meadowood immigrants or visitors, but communities with local traditions participating to the extent possible in the Meadowood phenomenon involving pan-regional, overlapping networks of alliances.  The artifacts most likely passed through a series of exchanges en route to the Delaware valley.   I thus see a certain interesting comparison between the Delaware valley sites and the Esmond 2 site, as the sites from these different regions seem to show evidence of differential participation in the Meadowood phenomenon, moderated by relative distance from sources of Onondaga chert.  Esmond 2 seems to have been a producer of objects exchanged in this network of alliance building; the Delaware valley sites appear to be on the periphery of the larger Meadowood network.

We are grateful to two students from Skidmore College, our lab/office assistant Priscilla Montalto and our intern Andrew Bosworth for help this semester as we processed and analyzed data from the Fernlea 1, Fernlea 2 and Esmond 2 sites.


Kinsey, W. Fred III
1972    Faucett Site, 36-Pi-13A.  In Archeology in the Upper Delaware Valley, edited by W. Fred Kinsey, III, pp. 159-197.  Anthropological Series No. 2, The Pennsylvania Historical and Museum Commission, Harrisburg.

Kraft, Herbert C.
1975    The Archaeology of the Tocks Island Area.  Archaeological Research
Center, Seton Hall University Museum, South Orange, New Jersey.

Tache, Karine 
2011    New Perspectives on Meadowood Trade Items.  American Antiquity

Versaggi, Nina M., LouAnn Wurst, T. Cregg Madrigal, and Andrea Lain
2001    Adding Complexity to Late Archaic Research in the Northeastern Appalachians.  In Archaeology of the Appalachian Highlands, edited by Lynne P. Sullivan and Susan C. Prezzano, pp. 121-133.  The University of Tennessee Press, Knoxville.

Werner, David J.
1972    The Zimmerman Site, 36-Pi-14.  In Archeology in the Upper Delaware Valley, edited by W. Fred Kinsey, III, pp. 55-130.  Anthropological Series No. 2, The Pennsylvania Historical and Museum Commission, Harrisburg.