«By James R. Keron Graduate Program in Anthropology Submitted in partial fulfillment of the requirements for the degree of Master of Arts Faculty of ...»
Third, despite some misgivings, another table was produced with the Coefficients of Similarity for the chert source types for all pairs of sites. Again any site with less than 100 pieces of debitage was dropped. The resulting table was then scanned for divergent pairs of sites. A lower value was used here as in general the coefficients were much lower than in the preceding table. Given the distance decay of Kettle Point chert and the dramatic changes between periods the only differences deemed worthy of consideration were between pairs of sites in the same time period and occurring close to each other.
This table is included in Appendix D as Table D-14.
At the end of these three analytical procedures, sites identified that recurred
frequently in each analysis were identified and are discussed individually here.
Melbourne-7 (AfHj-17) This Glen Meyer site stands out primarily with respect to chert source use. Use of Onondaga chert at 84% is extremely high and occurs at the expense of both Kettle Point chert and local till chert. The lithic industry as evidenced by the frequency of the various flake types is largely consistent with the other Glen Meyer sites since the coefficients of similarity between this site and the others are in excess of 160 for all but two other Glen Meyer sites. The site is represented by only a surface collection so any interpretations are problematic. However, the fact that it is high in Onondaga and is the most southerly site in the Caradoc cluster, and consequently closest to Lake Erie where it is presumed that Onondaga chert was acquired, would suggest that it may be a way station used by people recently returning from an Onondaga chert acquisition trip. This situation may be similar to the excavated McGrath site which was interpreted by Poulton (1985a) as a small temporary location as opposed to a cabin site or village. McGrath had a very high incidence of Kettle Point chert in the debitage suggestive of a recent acquisition trip.
Dorchester (AfHg-24) This site is a large Uren village at the eastern end of the study area. The primary difference with this site is again chert source frequencies. This site has a high percentage of Onondaga chert in the debitage at 53.6%. Kettle Point is lower than other MOI sites but it is the furthest MOI site from the source. While, the use of local till chert is also lower than for other period sites there is much more of it with cortex, the percentage is
19.3 higher than the average of all MOI sites. The percentage of shatter is also higher.
Both of these occur at the expense of core trimming flakes where the percentage is 15.3 lower than average. This local till chert is not used as frequently as other period sites but the till chert that is used is arriving in a less reduced state. Outside of the differences within local till chert use the overall lithic assemblage from Dorchester is not unusual.
Sifton (AfHh-85) This site has been fully excavated and has been interpreted by Pearce (1996) as a Uren cabin site. The key difference with this site is that there is a very low percentage of high quality chert. Onondaga is 2.5% and Kettle Point is 13.5%. These low frequencies are offset by much greater use of local till chert. While seemingly anomalous, it should be pointed out that in the sample analyzed, this site is the only MOI identified as an agricultural cabin. As note above, during Neutral times cabin sites show a higher frequency of use of local till chert. Within the Onondaga chert sample there seems to be a higher use of core trimming flakes but the sample is small. The overall chert working is in line with other MOI sites and, where different, varies as a Neutral cabin site would from a village.
Norton (AfHh-86) This is a village site that was partially excavated during a mitigation of a water main (Cooper and Robertson, 1993). The excavation is thus long and narrow through the village and it transected a number of houses. As it relates to chert sources used it is not substantially different than other MOI sites although use of Kettle Point chert is a bit lower and compensated for by the greater use of Onondaga chert. However, the lithic industry is somewhat different than in other MOI sites showing lower coefficients of similarity with three of the six other sites. In general there seems to be less emphasis on core trimming flakes and more on bifacial retouch and fragmentary flakes. These latter two categories though do not seem to be related. The major differences occur with Kettle Point and local till chert, where there seems to be more fragmentary flakes produced at the expense of core trimming flakes. The increase in bifacial retouch flakes though occurs with Onondaga chert which has a percentage that is 16.3 higher than the period average. The increase in Onondaga bifacial retouch flakes comes at the expense of core trimming flakes again. This difference would seem to imply a shift in the reduction strategy employed at the Norton site but how and why it differs is problematic. Given the internal spatial analysis conducted above, the general conclusion was that no major internal spatial distinctions were observed so it is unlikely that it is due to partial excavation of the site.
Laidlaw (AfHh-1) This site is interpreted here as a Neutral cabin site. It has been termed a village elsewhere (e.g. Pearce 1996) but this characterization has never been established as the adjacent field was not cultivated until quite recently. Although it had been intended to conduct a CSP on the site as part of the research for this thesis, such did not happen due to a combination crop cover, other time constraints and the inability to contact the owner to get permission. The sample used was taken from a midden only excavation conducted by Dr. Wm. Finlayson in 1974. As with the Norton site, there are some differences in access to chert. The site has a much lower percentage of Kettle Point chert at 5.5% of the total. This amount contrasts sharply with the Skinner site (AfHg-13) less than 200m away on the other side of the creek which has 24.4% Kettle Point chert. However, there are a number of differences in the reduction sequence. In considering the coefficients of similarity with respect to the flake types, it is less than 160 for four of the other eight Neutral sites. The major difference seems to be a lack of fragmentary flakes. This result can be found in the total and severally in each of the chert types excluding Kettle Point which is too rare to consider. Use of local till chert is higher and primary reduction is higher (decortication flakes and shatter) and use of high quality cherts is less. Both of these trends would be typical of a Neutral cabin site.
With the Skinner site nearby, and both samples being from excavated middens, the opportunity for comparing these seems worth following up. Accordingly, the differences were computed in a similar fashion to that used to compare to period averages. The first point here is that the differences in chert source usage do not appear to be out of line with differences to other Neutral sites despite the difference in the use of Kettle Point chert. The second point is that the Skinner sample is much larger than the sample from Laidlaw (total debitage of 806 versus 218) despite the fact that it was taken from only 5 square metres of midden. In looking at these two sites a similar trend in comparison to site average is noted. There is a much lower frequency of fragmentary flakes at Laidlaw but a much higher frequency of core trimming flakes. Other than concluding that there was a difference in functions performed at the two sites despite their close proximity, any further explanation is problematic.
Chapter 7: Evaluation of Hypotheses A set of behaviourial hypotheses and potential implications of the same was developed in Chapter 3 and the data analysis yielded a number of observations in Chapters 5 and 6. In order to evaluate the success or failure of the hypotheses, the observations were mapped against the implications and hypotheses. These are included in Appendix E. The following discussion addresses each of the hypotheses in turn considering both the evidence obtained here as well as other evidence and trends both in Iroquoian studies and beyond that bear on the various questions. In general there was good support from the data for most of the hypotheses with the notable exception of the time depth of the inferred down-the-line exchange mechanism. This notion was not supported and what emerged from the data was an acquisition pattern during Glen Meyer times that was significantly different from the pattern observed later during Neutral times.
The hypotheses are discussed in order.
Hypothesis 1: Kettle Point and Onondaga Were Preferred In early formulations, this idea was subsumed as part of the down-the-line hypothesis as a result of understanding the knapping properties of the various chert sources used within the study area. Later it was realized that it may well be a behavioural hypothesis on its own. However, as the research design was already established and data collected, it did not seem worthwhile to go back and establish testable implications when the data had already been collected. Nonetheless, in reviewing the observations several of these tended to provide verification of this hypothesis. These appear in Appendix E, Box 1.
Some of these indications derive from distribution of chert types within a village, for example at the Brian site (AfHh-10) the external midden is low in Onondaga chert and while not demonstrable through the data used here, it is also low in Kettle Point chert. Also local till chert is more prone to end up in the middens indicating a lesser value. Some areas within sites also have a very low level of local till chert and a correspondingly higher level of Kettle Point and Onondaga chert indicating that it would receive preferential use if it can be obtained. Another major indicator is the level of use in the late prehistoric Neutral time frame. At Lawson (AgHg-1), and in the cluster in the Lambeth vicinity, there is a very high usage of Kettle Point chert despite the fact that there is a supply of local till chert that seems to have been useful, especially during the MOI time. Finally, during the MOI when access to Kettle Point chert is curtailed, there is an increase in use of local till chert for projectile point manufacture as there is no high quality alternative.
All of this demonstrates that the high quality chert is preferable if it can be obtained. Local till chert is something that is used if nothing else is available.
Hypothesis 2: Down-the-Line Exchange As noted above, the time depth hypothesis failed to hold up and while it is discussed below in detail, the implications of this fact must be taken into account in this evaluation. The discussion will commence with the early historical records and then proceed back through time considering each the three time periods looking at the evidence from the study area as well as from other sources that bear directly on these questions.
The Antiquty of Exchange While Iroquoian trade and the control of trade routes by lineages is unequivocal in the early historical records, Trigger (1987: 168) notes there was some controversy at the time of his writing (the first edition was in 1976) both with respect to the antiquity of trading relationships and with lineage control of trade routes. Part of the problem is undoubtably the lack of clearly identifiable exotic goods originating at substantial distances from the point of deposition as is evident during Late Archaic and Middle Woodland times. These items are very rare on Iroquoian sites of any time period.
Furthermore, as Trigger notes, in the historic period with the intensification of trade driven by the fur trade and European goods, items from some distance away again become evident in the archaeological record. However, he notes that even in historic time Athe bulk of the trade appears to be in perishable goods, such as corn, fish, and skins@ (Trigger 1987:169). These items plus other material such as nets and rope are not normally preserved or if they are, it can be difficult to tell whether or not it was as a result of exchange that lead to the final deposition. He then uses the occurrence of Iroquoian pottery throughout much of the Algonkian territory through all periods of the Late Woodland as the basis to conclude that trade was present prehistorically and that the French only tapped into a complete functional system of trade with their demand for furs in exchange for European goods. Throughout much of the rest of the book it is quite evident that the Native Americans were not newcomers to trading but had a complete set of rules of interaction and a strategy of controlling access by one=s downstream partners to ones upstream partners.
Subsequent work has confirmed Trigger=s assessment of the antiquity of exchange patterns. As noted above, Iroquoian lithic analysis (Fox 1990) has shown that exchange of chert nodules and finished artefacts between major tribal groups was occurring in late prehistory. Recently, an edited book (Baugh and Ericson 1994) has been published containing a series of regional syntheses that demonstrate an ongoing system of regional exchange extending back several thousand years in most areas of the continent. Brose (1994) includes the Great Lakes drainage in an area he calls the AMid West@ and describes a pattern where after the collapse of the Middle Woodland exchange systems, there is a period where the amount of exchange drops significantly during the early Late Woodland and the patterns of exchange of raw materials Asuggest focused individual or family initiative rather than down-the-line exchange@ (Brose 1994: 228). Trade withers to a point where only Lake Superior copper and Atlantic shell products are exchanged through Ontario. Following AD 1400, there is a resurgence in exchange where Alimited lineagecontrolled, family organized trade in subsistence resources with surrounding groups@ occurred (Brose 1994: 230).
In general, since Trigger=s (1987) writing in 1976, the proposition that prehistoric groups through the Late Woodland period in southern Ontario engaged in various forms of exchange with neighboring groups has been well established and the only debate now
centres on the nature and extent of the exchange.