Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America

JASREP-00671; No of Pages 8 Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep

Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America Nathaniel R. Kitchel Department of Anthropology, University of Wyoming, 1000 E. University Ave., Laramie, WY 82071; United States

a r t i c l e

i n f o

Article history: Received 19 October 2015 Received in revised form 13 October 2016 Accepted 17 October 2016 Available online xxxx Keywords: Colonization Paleoindians Hunter-gatherers X-ray fluorescence Geochemical sourcing Northeastern North America

a b s t r a c t In this study Energy Dispersive X-Ray Fluorescence (EDXRF) is used to confirm previous visual identifications of red Munsungun chert, in 23 fluted point sites, or loci, in New England and southern Quebec. These source designations are combined with previous visual raw material identifications to provide a proxy measure of landscape knowledge possessed by early inhabitants of the region. The results herein show these groups possessed significant landscape knowledge and patterned lithic raw material procurement strategies. These results fail to support the idea that some behavioral adaptations during the early fluted point period such as robust toolkit design, long distance lithic transport or other such behaviors are the result of landscape unfamiliarity. These patterns likely hold for other are colonized by mobile hunting and gathering groups. © 2016 Published by Elsevier Ltd.

1. Introduction Clovis and other early fluted point technologies are the first widespread lithic technology in the Americas. If these technologies represent the very first populations in the New World remains a matter of debate in American archaeology (Adovasio et al., 1990; Dillehay, 1989, 1999, 2000; Gilbert et al., 2009, 2008; Meltzer, 2002: 30; Waters and Stafford, 2007; Waters et al., 2011a; Waters et al., 2011b). Debate also surrounds the duration of the Clovis period, proposed to last as little as 250 years (Waters and Stafford, 2007), or possibly over 1000 (Prasciunas and Surovell, 2015). In the event Clovis does not represent the first populations in the Americas, or the Clovis period lasted significantly longer than the 250 years proposed by Waters and Stafford (2007) it is unlikely that all Clovis sites are representative of the colonization, or the landscape learning process (e.g. Ellis and Lothrop, 1989: 149; Snow, 1980). Despite such uncertainty, archaeologists continue to attribute ostensibly unique behaviors in the Clovis record, such as long distance lithic transport and robust toolkit designs as adaptations to landscape unfamiliarity (e.g. Andrews et al., 2015; Boulanger et al., 2015; Meltzer, 2009: 252; Tankersley, 1991: 297). The research herein uses lithic raw material procurement patterns amongst early fluted point groups in northeastern North America as a proxy measure of landscape knowledge. Visual identifications of lithic raw materials were augmented by XRF geochemistry. No evidence was found to indicate that these early populations had limited knowledge of their landscape.

E-mail address: [email protected].

The Northeast (Massachusetts, Maine, New Hampshire, southern Quebec, and Vermont for the purposes of this study) is an ideal location to test colonization models because of a well understood deglaciation chronology (Ridge et al., 2012), a small number of lithic raw material sources with both large and small source areas (see Burke et al., 2014; Table 1), and a number of excavated fluted point sites located across a large geographic area (Spiess et al., 1998). Glacial ice rendered far northern New England unavailable to human occupation just prior to the appearance of early fluted point technology in other areas of North America (Sanchez et al., 2014, see also Ferring, 2001: 50). This ice precludes the possibility of an earlier “pre-Clovis” occupation in the region rendering a lengthy, but archaeologically undetected exploration period impossible in this region. Though “Clovis” technology sensu stricto is not found in the Northeast, three point types collectively placed into the “early fluted point period” of the region (Bradley et al., 2008) are often argued to be colonizing populations (Dincauze, 1996: 10). Ice also covered five separate lithic raw material sources in northern New Hampshire and Maine. These sources are Jefferson rhyolite, Ledge Ridge chert, Mount Jasper rhyolite, and most importantly Munsungun chert (Table 1). Ice blanketed a number of small chert sources in southeastern Québec (Burke, 2007; Pollock et al., 2008; Pollock et al., 1999: 281), though these have not been identified in any fluted point site lithic assemblages to date. In the Lake Champlain Valley of Vermont, the bedrock source for Hathaway and similar cherts (Burke, 1997: 44–46) remained buried by ice upon the first appearance of early fluted point technology in North America and were then likely inundated by the Champlain Sea following glacial retreat (Robinson, 2012). Larger

http://dx.doi.org/10.1016/j.jasrep.2016.10.009 2352-409X/© 2016 Published by Elsevier Ltd.

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

2

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Table 1 Lithic source descriptions. Name

Stone type

Cheshire quartzite

Metaquartzite N100 km2

Location

Secondary transport

Description

Citations

Unknown

White vitreous quartzite

Condon (1993) and Ratcliffe et al. (1975)

b100 m2

Northwest Vermont to Connecticut Berlin, NH

Mt. Jasper rhyolite Jefferson rhyolite

Flow Banded Rhyolite Flow banded rhyolite

None

Unknown

Jefferson, NH

Boisvert (1992) and Pollock et al. (2008) Pollock et al. (2008)

Mooshead Lake, ME

Olive green weathering to white. Frequent phenocrysts. Very homogenous.

Rankin and Caldwell (2010) and Bourque (1995: 102)

Chert

≈80 km3 16,000 km2 (includes secondary deposits) 100 m2

Limited to low densities in gravel deposits near Jefferson, NH Found in deposits from the source southeast to the seacoast

Olive green to brownish weathering to nearly white Pink, olive yellow, blackish green. Coarse to microcrystalline in texture

Northwest, ME

Unknown. Likely very limited

Green to black. Some pyrite rhombs

Chert

10 km2

Munsungun Lake, ME

Limited

Black to light gray, green and red.

Gramly (1982) and (Gramly (2009: 124) Hall (1970), Pollock (1987) and Pollock et al. (1999)

Kineo/Traveler Rhyolite rhylite

Ledge ridge chert Red Munsungun chert

Source size

bedrock exposures and earlier deglaciation led to the availability of other raw materials, such as Cheshire quartzite and Kineo/Traveler rhyolite (Table 1) prior to the availability of smaller more northerly sources. Glacial transport of material from these larger sources is much more significant than from smaller sources further extending the spatial area over which these sources are available (Fig. 1). Of the five lithic sources used for this study, three lie to the north of the ice margin circa 13,500 cal B.P. (Fig. 1), meaning that even if the

region was settled at this time these sources were not available. These raw material sources, Jefferson Rhyolite, Mt. Jasper Rhyolite, and Munsungun Chert, are also found over limited spatial areas (Table 1). Due to the relatively small size of their bedrock outcrops these materials have limited natural transport of the material from their primary source area. Like the materials discussed by Tankersley (1989: 261), these materials are not available in meaningful quantities or sizes in secondary deposits distant from their bedrock source. Thus, the possibility of

Fig. 1. Map shows locations of archaeological sites and lithic sources in the Northeast. Approximate ice margin locations at 14,000 to 13,800 and 13,700 to 13,500 Cal. B.P. are also indicated. Ice margins approximated from Ridge et al. (2012).

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

discovery of the primary outcrop from secondary deposits is extremely limited. 1.1. Expected signature of a colonizing group It is expected that colonizing groups lacking landscape knowledge (Kelly and Todd, 1988) will encounter lithic sources with the largest spatial extents (e.g. Kineo/Traveler) early in the settlement process based solely on the higher probability of encountering such large sources while traversing the landscape. Here, spatial extent refers to both the area of the bedrock source itself, as well as to the extent of secondary deposits of the material. A lithic source may also be preferentially encountered and used, regardless of size, if the source is located along a potential travel, or exploration, corridor such as a large river (Anderson and Gillam, 2000). Lithic materials from previously settled areas outside the region may also be expected to be used extensively or exclusively during a colonization event where an incipient population lacked knowledge of lithic raw materials in the region. To examine whether lithic raw material assemblages from early fluted point sites in the Northeast followed the expected pattern of raw material use during a colonization pulse, previous lithic raw material identifications were compiled from 23 early and middle fluted point sites (Bradley et al., 2008), or loci in the region (Tables 2 and 3). Red Munsungun chert, found in the Munsungun Lake formation, Piscataquis County Maine, was of particular interest to this study because it is the most northerly and recently deglaciated source and commonly thought to be present in many fluted point sites in the Northeast. To support the visual source identifications of raw materials, which are sometimes inaccurate, non-destructive Energy Dispersive X-Ray Fluorescence (XRF) was used to evaluate the geochemical homogeneity of red chert artifacts visually identified as red Munsungun chert. These source attributions were then used to evaluate whether the presence or absence of specific raw materials of early fluted point sites in the region demonstrate lithic use patterns consistent with limited of landscape knowledge expected during a colonization pulse (e.g. Andrews et al., 2015; Boulanger et al., 2015: 555; Kelly and Todd, 1988). All early fluted point sites in the region demonstrate a consistent pattern of the use of a small suite of high quality lithic raw materials regardless of the size or geographic location of the source. Several sources found over the largest spatial areas are used less frequently, while red Munsungun chert was ubiquitous, suggesting a highly patterned, rather than opportunistic lithic procurement strategy. These patterns do not suggest landscape unfamiliarity was a primary driver in lithic raw material procurement strategies during the early fluted point period. This indicates learning events amongst highly mobile hunter-gatherers can be archaeologically invisible. 1.2. On finding rock Prehistoric peoples probably quickly recognized certain rock types often contain raw materials suitable for the manufacture of stone tools (e.g. chert in limestone). This knowledge is particularly useful when

3

these rock types appear over large areas on the landscape and are exposed by erosion, which is not the case in the Northeast. In this region, geologic formations are often convoluted with a great deal of metamorphic and igneous rock and few unaltered sedimentary rocks. Therefore, prehistoric peoples could not have relied on previous experiences from other regions to predict or detect suitable toolstone. For example, it is highly unlikely that a group of colonizing foragers could predict the location of knappable sized joint blocks of high quality red Munsungun chert within a majority of igneous and metamorphic rocks composing the Munsungun Lake Formation (Hall, 1970). Even more difficult to predict through geologic knowledge alone are the New Hampshire rhyolites. Only two of the dykes mapped in the north country of New Hampshire are known to yield knappable stone (Billings and Fowler-Billings, 1975). These dykes do not follow predictable patterns in regard to their location on the landscape or their relationship to the local bedrock. In this geologic environment colonizing groups would be forced to find toolstone sources by direct encounter rather than relying on previous geologic knowledge. Though Paleoindian groups in the Northeast preferred to use stone derived from bedrock locations (Gardner, 1989), secondary deposit size must be considered because these secondary deposits could have led foragers to the bedrock location of these materials. Alibates silicified dolomite is one relatively well known example. Although the exposed bedrock source for Alibates is relatively small, knappable nodules of the material can be found in gravel deposits of the Canadian River hundreds of kilometers downstream (Wyckoff, 1993). Witthoft (1952: 471) also notes glacially transported Onondaga chert cobbles at the mouth of the Susquehanna River at the Chesapeake Bay. Prehistoric peoples could procure the materials directly from the secondary (riverine or glacial) deposit itself or by following the cobble train back to the bedrock source (Meltzer, 1984). Of the Northeastern materials only Kineo rhyolite is known to have been naturally transported large distances in significant quantities from the bedrock source. Cheshire quartzite may also have been transported into the Champlain Valley, by glacial or fluvial action, though this is currently unevaluated. Conversely, many of the other sources are difficult to detect even when virtually standing on the source itself, such as Mt. Jasper rhyolite, or the small quarry location for red Munsungun chert (Kitchel and Barker, 2016). In the Northeast, sources such as Munsungun represent point locations on the landscape. Knowledge of these locations demonstrates knowledge of a specific and difficult to detect patchy resource (see Rockman, 2003). The presence of stone from these small sources in archaeological sites allow an evaluation of the speed of learning and acquisition of local (point) knowledge through time, as each raw material source location entered the overall knowledge base of a colonizing group. If the landscape learning process was visible in the archaeological record, sources with larger primary or secondary source areas will appear in the archaeological record before smaller sources. Alternatively, the presence of lithic materials in early Paleoindian toolkits could be time transgressive from south to north if colonizing groups moved from a southerly entry points to the north encountering lithic sources

Table 2 Raw material counts from sites or loci associated exclusively with early fluted point diagnostics (Bradley et al., 2008). (See Boisvert, 2012, 1999, 1998; Cox, 1998; Cox et al., 1994; Crock and Robinson, 2012: 51–52; Curran, 1994, 1984; Gramly, 2010, 2005a, 2005b, 1995, 1984, 1982; Gramly and Rutledge, 1981; Gramly and Storck, 1988; Goodby et al., 2011 for site descriptions.) Site name Raw material

Vail

Morss

Adkins

Searsmont

Mahan

Tenant Swamp

Whipple

27-CO-29 (block y)

Kineo/Traveler Cheshire quartzite Munsungun Hudson Valley New Hampshire rhyolite Pennsylvania Jasper Quartz crystal Unknown

– – 29 10,272 – 474 14 875

– – 414 – – – – 1

– 6 24 – – 1 167 156

220 1 105 – 339 – 193 621

– 325 67 Present 2 31 14 701

– – 2172 – 2448 – 571 16

– – 54 24 3 4 15 67

– 7 53 – 2155 – 36 40

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

4

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Table 3 Raw material counts from sites or loci associated exclusively with middle fluted point period (Michaud-Neponset) diagnostics (Bradley et al., 2008). (See Boisvert, 2012, 1998; Burke et al., 2014; Carty and Spiess, 1992; Chapdelaine, 2012; Crock and Robinson, 2012; Donta, 2005; Spiess et al., 1998; Spiess and Wilson, 1987; Wilson, 2003 for site descriptions.) Site name Raw material

Cliche-Rancourt 27-CO-28 27-CO-30 (Locus 3)

Jackson-Gore Dam Lamoreau Lautman Michaud Colebrook 27-CO-29 (block X)

Neponset

Kineo/Traveler Cheshire quartzite Munsungun Hudson Valley New Hampshire rhyolite Pennsylvania Jasper Quartz crystal Unknown

– 6 1287 – 83

– – 1 – 171

– 1 133 – 3178

– 1 261 26 –

– 32 141 150 6

– 1 1709 – 1262

7 1 142 – 7

– 1 649 546 859

– – 265 – 193

3 – 13 – 2264

– 5 186 – 988

– 237 51

– 21 2

– 139 11

8 8 826

– 29 208

– 2 6

– – 23

– – 227

– 129 41

– 12 17

16 – 507

in the process. A similar pattern would manifest if colonizing groups moved northward following retreating glacial ice. In these scenarios materials from the south or west of the region will appear in early fluted point lithic assemblages before more northerly sources. 1.3. Geochemistry The XRF instrument used for this analysis was a PANalytical 4.0 kW sequential wavelength dispersive spectrometer system equipped with a 4 kWrh end window ZETA technology x-ray tube. The instrument also has 5 analyzer crystals (two curved), three collimators, programmable six collimator mask and three flow and sealed proportional in-tandem detectors and scintillation detector in parallel. This instrument can detect any element lighter than fluorine to the low parts per million level (though the exact precision depends on the atomic mass of the element in question as well as the surface area of the analyzed sample). 740 artifacts of suitable size for this instrument were selected from 23 fluted point sites for analysis. Over 235 geologic samples were also collected from primary bedrock contexts at locations showing evidence of prehistoric quarrying. Following Gauthier et al. (2012) comparisons of elemental and major oxide concentrations amongst these objects were made using simple bi-plots and Principal Component Analysis (PCA), implemented in JMP©, Version 10. Though numerous material types were analyzed, the following discussion focuses on the confirmation of visual identifications of red Munsungun chert (Fig. 2). The analysis of all unknown or visually indistinct materials is currently ongoing and not reported herein. Geochemically, all artifacts visually identified as red Munsungun chert show strong homogeneity between sites, demonstrating that the visual identifications of this material are internally consistent. For example Fig.

Fig. 2. A bedrock outcrop of red Munsungun chert associated with prehistoric quarrying activities.

3 shows differences in calcium and manganese concentrations between artifacts visually identified as red Munsungun Chert and items visually identified as Hudson Valley Chert. This trend of internal geochemical homogeneity for artifacts visually identified as red Munsungun chert is also evident in a Principal Component Analysis (PCA) using concentrations of Rb, K, Ba, Sr, Mn, La, Ce, Y, Th, Zr, and Nb in the entire sample of XRF analyzed red Munsungun items (Fig. 4). Here principle components one and two display a positive linear relationship for red Munsungun, while Hudson Valley materials lack this relationship. A PCA using the same elements as above examining only artifacts visually identified as red Munsungun chert from eight early and middle fluted point sites, shows a lack of clustering of artifacts by site. These results indicate that each of these items was produced of chert from the same or closely related outcrops (Fig. 5). Unfortunately as Burke et al. (2014: 123) have noted, geoarchaeological prospecting in the vicinity of Munsungun Lake is ongoing and the identification of outcrop locations for red Munsungun chert outcrops associated with quarrying activities has not been straightforward. Because of these factors the only documented outcrop of red Munsungun chert associated with prehistoric quarrying activities was discovered in the summer of 2015 (Kitchel and Barker, 2016). Samples taken from this location are currently submitted for XRF analysis, but these analyses are not complete. Despite the limitations of the geochemistry, previous petrographic analyses have linked some of the red chert artifacts analyzed for this study to the Munsungun Lake Formation. As Fig. 5 shows, the analysis described above included artifacts of red chert from the Morss fluted point site located in western Maine just east of the New Hampshire boarder (Fig. 1). Petrographic analysis of red chert items from this site

Fig. 3. Bivariate plot showing Ca versus Mn concentrations for artifacts visually identified as Hudson Valley and red Munsungun cherts. Elemental concentrations have been normalized to a mean of zero.

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

Fig. 4. Principal Component Analysis (PCA) analysis of concentrations of Rb, K, Ba, Sr, Mn, La, Ce, Y, Th, Zr, and Nb for items visually identified as Hudson Valley and red Munsungun cherts. 95% density ellipses are shown.

show this material to be consistent with chert from the Munsungun Lake formation (Pollock et al., 1999). Additionally, these items clearly fall within the geochemical distribution of items identified as red Munsungun chert from other sites analyzed in this study (Fig. 5). The similar geochemistries of items visually identified as Munsungun Chert confirm that the macroscopic visual identifications of red cherts from multiple archaeological sites have internal geochemical consistency. This internal consistency and lack of clustering show that these visual source attributions are reliable between sites. Nonetheless, until the additional geochemical analyses mentioned above are completed, linking archaeological artifacts to a specific quarry or outcrop location within the Munsungun Lake Formation remains impossible. 1.4. Raw material use during the fluted point period No early fluted point assemblages in the Northeast consist exclusively of materials from the south or west, as would be expected if

Fig. 5. A Principal Component Analysis (PCA) using concentrations of Rb, K, Ba, Sr, Mn, La, Ce, Y, Th, Zr, and Nb for all artifacts visually identified as red Munsungun chert. 95% density ellipse is shown.

5

colonizing groups were reliant on raw materials from previously settled regions. Nor do these assemblages consist entirely of materials derived from the largest lithic sources in the region as would be expected if early fluted point groups lacked detailed landscape knowledge. Rather, raw materials found over the largest areas tend to occur at low frequencies in the lithic raw material assemblages of early fluted point sites in the Northeast (Table 2). Though Cheshire quartzite is present in many of these assemblages, it occurs at low frequencies. Kineo/Traveler rhyolite occurs over the largest area yet appears only infrequently and in small quantities in early fluted point sites (Table 2). The majority of fluted point sites that do contain Kineo/Traveler rhyolite are most often found located on secondary deposits of the material (e.g. Michaud, a middle fluted point site [Spiess and Wilson, 1987]), indicating that this material was used only opportunistically. Early fluted point assemblages more often contain lithic materials from a suite of smaller high quality sources, rather than the large lithic raw material sources expected if these early groups were unfamiliar with available lithic resources (Table 2). With the exception of Vail (Gramly, 2010) and Whipple (Curran, 1994, 1984), lithic raw materials in early fluted point sites do not predominantly come from sources located to the south or west of the study area as would be expected if source acquisition was time transgressive from south to north, following population movements or glacial retreat. Initially the presence of a majority of Hudson Valley chert in the Vail and Whipple assemblages indicate the inhabitants of these sites lacked knowledge of northerly lithic sources, but the presence of small quantities of red Munsungun chert in each of these assemblages contradicts this conclusion. The presence of red Munsungun chert demonstrates that these early groups had already traveled to the most northerly and recently deglaciated portion of the study area, discovered the source of this material, and returned south. Given the presence of red Munsungun chert in the assemblages of these sites the lithic raw material ratios found at Vail and Whipple likely arise from the vagaries of seasonal rounds and lithic procurement strategies and discard patterns, not unfamiliarity with the landscape. Other early fluted point assemblages in the region generally contain a mixture of red Munsungun chert, Hudson Valley chert, New Hampshire rhyolite and Pennsylvania jasper, along with other often unknown materials. Raw material ratios in fluted point sites from the Northeast are highly variable (Table 2) with many materials appearing only occasionally, particularly distinct but unknown materials. The one notable exception is red Munsungun chert which is found not only in all early fluted point sites, but is present in every fluted point site examined for this study, sometimes making up the vast majority of a site assemblage (Tables 2 and 3). In addition to Munsungun chert, early lithic assemblages in New England area sometimes contain New Hampshire rhyolites, (e.g. Robinson et al., 2009: 427). Several early fluted point loci have also been located directly upon till deposits that contain Jefferson rhyolite. These single component loci variously contain all early fluted point period forms (Boisvert, 2012; Bradley et al., 2008), demonstrating knowledge of these sources amongst some of the earliest inhabitants of the region. Rather than having limited knowledge, these consistent lithic procurement patterns demonstrate that early groups exploited many high quality lithic raw materials in the region regardless of source size or location. These findings are contrary to the expectations that early groups lacked landscape knowledge. That the only material found in all early fluted assemblages, red Munsungun chert, arises from the northernmost, and most recently deglaciated, portion of the study area demonstrates significant landscape knowledge. More telling is that the red variety of Munsungun chert was most frequently used variety of Munsungun chert, rather than the more common, and easily detected, gray and black varieties. Not only is the red variety of Munsungun Chert the least common colour of chert in the Munsungun Lake Formation, outcrops where joint blocks of this material are suitable for the production of stone tools are rare, with only one location currently

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

6

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

known (Kitchel and Barker, 2016). Though the proportion of this material varies greatly between sites, its ubiquity demonstrates that this source was well known to early populations who favored the red variety of this material over other types. These source data also indicate the existence of patterned and extremely large lithic conveyance or transport zones (sensu Jones et al., 2003; Smith, 2010). Four of the six early fluted point assemblages in this study contain lithic materials visually identical to materials from Pennsylvania and Hudson Valley sources as well Munsungun chert. Such consistent lithic transport patterns are also contrary to the hypothesis that early groups had yet to settle into the landscape (Kelly and Todd, 1988). Instead, they point to the existence of well-developed, highly structured, landscape movement patterns and social connections over large distances amongst the earliest detected inhabitants of the region. Lithic raw materials are also transported primarily from the southwest to the northwest, and vice versa, following the Appalachian spine (see Ellis, 2011), with little evidence of a long distance pattern of west to east transport of materials, which some have argued is indicative of a colonization event (Tankersley, 1991: 297). These results are congruent with those of Ellis (2011), and again indicate the early fluted point record of northern New England does not show a pattern constant with landscape unfamiliarity or time transgressive population movements. Ellis (2011: 387) has argued that these large magnitude movements indicate these populations were “closer” to the colonization process, based on the belief that population densities were very low at this time, though how population density relates specifically to colonization or “settling in” is not explicitly discussed. The data presented here strongly indicate that early fluted point groups in the Northeast did not lack landscape knowledge, but were rather already embedded on the landscape at the time these populations became detectable in the archaeological record. These groups possessed landscape knowledge spanning a large geographic area indicated by the ubiquity of red Munsungun chert in all early fluted point sites. By the time early populations in the region became archaeologically visible, these groups had sufficient time to locate the majority of the high quality, or desirable (e.g. Goodyear, 1979), lithic resources in the region and to integrate these resources into patterned seasonal rounds, or logistical forays. Hazelwood and Steele (2003) discuss the possibility of such an invisible colonization event, arguing that until a population reaches a sufficient size this population will remain undetectable in the archaeological record. It appears that this is the situation in the Northeast where population size remained small enough for a sufficiently long period of time for these groups to learn their landscape while creating so few sites as to be functionally invisible in the archaeological record. 2. Conclusions Collectively, these findings fail to support the idea that the colonization, or more accurately, the landscape learning process of hunter gatherer populations is visible in the archaeological record. Rather this evidence shows early fluted point populations in the Northeast were not actively involved in the landscape learning process, but were already embedded on the landscape by the time these populations became archaeologically visible. This is demonstrated by the habitual use of lithic raw materials from relatively small sources while larger sources were used infrequently or ignored. The ubiquity of red Munsungun chert in all early fluted point sites in this study indicates the presence of highly patterned lithic procurement strategies during this time period, a virtually impossible occurrence if these groups were unfamiliar with their landscape. This outcome seems more probable when one takes into account how quickly the landscape learning process could progress as demonstrated by a simple model. The land area of the states of Maine, Vermont, Massachusetts, and New Hampshire is approximately 168,122 km2. Assuming a stable population of 100 foragers who explored an average of

0.25 km2 of land area per day, this group could collectively explore 25 km2 per day. Dividing the land area of northern New England by 25 gives yields 6725 days or about 18.4 years. While this model is obviously a gross oversimplification, it does demonstrate the rapidity with which a small population can explore large areas. Hypothetically, a highly mobile hunting and gathering population could encounter every lithic raw material source in the Northeast in approximately one generation, a timeframe so brief as to make archaeological detection of these events all but impossible. This pattern of rapid landscape learning is not restricted to the glaciated Northeast. As Reuther et al. (2011: 282) note in regard to early obsidian procurement in Alaska, “…(T)he earliest known occupants of eastern Beringia appear to have developed a relatively extensive knowledge of local and distant lithic resources, contrary to assertions that the earliest colonizers had little knowledge of high-quality, exotic lithic resources.” This similar pattern of detailed landscape knowledge amongst the earliest inhabitants of the region indicates that such an “invisible” colonization is not unique to the Northeast, but is likely a widespread phenomenon to be expected throughout the Americas as well. These findings demonstrate that behaviors observed in the archaeological record of the early fluted point, or “Clovis” period throughout North America cannot be taken uncritically as representative of the colonization process. Regardless of whether Clovis or the progenitors thereof, were the first inhabitants of the Americas the archaeology of the early fluted point period is not that of populations unfamiliar with their landscape. While some may consider population growth and the “filling in” of empty space to represent a “settling in” or colonizing process (Ellis, 2011), I argue that this is likely not related to landscape learning and in no way relates to whether these early groups possessed limited knowledge of their landscape. More likely, the archaeology of the early fluted point period is that of low population densities and high mobility, as opposed to that of a population learning and settling into an unfamiliar landscape. These findings further problematize colonization models derived from the early fluted point period, as do discoveries indicating the presence of a human population in the Americas south of the ice sheet before roughly 13,500 cal. B.P. (Adovasio et al., 1990; Dillehay, 1989, 1999, 2000; Gilbert et al., 2008; Meltzer, 2002: 30; Waters and Stafford, 2007; Waters et al., 2011a; Waters et al., 2011b). Taken together the existing data from early fluted point sites indicate considerable landscape knowledge amongst these groups. While the archaeology of the early fluted point period does not necessarily represent the initial colonization pulse into the Americas, it is still the first widespread and well-studied archaeological complex in the New World. Therefore, while this period may not directly relate to the colonization process in terms of learning unfamiliar landscapes and environments, these groups inhabited a social and natural environment markedly different from ethnographically documented hunter gatherer groups. These differences make the study of early fluted point archaeology that much more important, as many questions remain regarding how human groups behave at very low population densities, or adapt to enormous and rapid environmental change. Until more data are available concerning the progenitors of fluted point technology, only through the study of the early fluted point period can we hope to understand the behaviors leading to the ubiquity of this technology in North America. Beyond North America, I propose these patterns of landscape knowledge acquisition extend to northerly areas of Europe and Asia as well as human groups entered these areas for the first time. As in the Northeast, in these areas it is likely that modern human hunter gatherer groups were also able to learn new landscapes with such rapidity as to be archaeologically invisible for significant periods of time. The detection of these colonization events would also be that much more difficult in these regions because of the greater time depth of human settlement in these areas. Thus, archaeologists should not invoke landscape unfamiliarity to explain the behaviors of early populations, but rather seek

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

to understand what other circumstances might give rise to such behaviors. This will lead to a better understanding of the social, technological, and ecological factors leading to the sometimes extreme behaviors observed amongst early northern hunter gatherers freeing our interpretations from ad hoc models based on the archaeology of the first archaeologically visible populations in the New World (e.g. Kelly and Todd, 1988), rather than the truly first population in the New World. Acknowledgments I thank Heather M. Rockwell, Melissa Murphy, and Richard Boisvert for comments on previous drafts of this paper, as well as those from three anonymous reviews whose insights greatly strengthened the final version of this paper. I also extend a special thanks to Adrian Burke for introducing me to the Munsungun Lake area, as well as accompanying me on my summer 2015 fieldwork. I owe Richard Boisvert, John Crock, Jess Robinson, Arthur Spiess, Bruce Borque, and R. Michael Gramly a deep debt of gratitude for allowing access to the collections analyzed for this research. I am deeply indebted to Norbert SwobodaColberg, the resident geochemist at the University of Wyoming for incredible inputs of time, both operating the XRF instrument as well as post processing data on hundreds of lithic samples. I also owe Robert L. Kelly many thanks for inspiring this research. This research was supported by National Science Foundation Doctoral Dissertation Improvement Grant #1342656. All errors are my responsibility alone. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jasrep.2016.10.009. References Adovasio, J.M., Donahue, J., Stuckenrath, R., 1990. The Meadowcroft Rockshelter radiocarbon chronology 1975–1990. Am. Antiq. 55 (2), 348–354. Anderson, D.G., Gillam, J.C., 2000. Paleoindian colonization of the Americas: implications from an examination of physiography, demography and artifact distribution. Am. Antiq. 65 (1), 43–66. Andrews, B.N., Knell, E.J., Eren, M.I., 2015. The three lives of a uniface. J. Archaeol. Sci. 54, 228–236. Billings, M.P., Fowler-Billings, K., 1975. Geology of the Gorham Quadrangle: New Hampshire-Maine. Department of Resources and Economic Development, State of New Hampshire. Boisvert, R., 1992. The Mount Jasper Lithic Source, Berlin, New Hampshire: national register of historic places nomination and commentary. Archaeol. East. N. Am. 20, 151–166. Boisvert, R., 1998. The Israel river complex: a Paleoindian manifestation in Jefferson. New Hampshire Archaeology of Eastern North America. 26, pp. 97–106. Boisvert, R., 1999. Paleoindian occupation of the White Mountains, New Hampshire. Géog. Phys. Quatern. 53 (1), 159–174. Boisvert, R., 2012. The Paleoindian period in New Hampshire. In: Chapdelaine, C. (Ed.), Late Pleistocene Archaeology & Ecology in the Far Northeast. Texas A&M University Press, College Station. Boulanger, M.T., Buchanan, B., O'Brien, M.J., Redmond, B.G., Glascock, M.D., Eren, M.I., 2015. Neutron activation analysis of 12,900-year-old stone artifacts confirms 450– 510+ km Clovis tool-stone acquisition at Paleo Crossing (33ME274), northeast Ohio, U.S.A. J. Archaeol. Sci. 53, 550–558. Bourque, B., 1995. Diversity and Complexity in Prehistoric Maritime Societies: A Gulf of Maine Perspective. Plenum Press, New York, NY. Bradley, J.W., Spiess, A.E., Boisvert, R.A., Boudreau, J., 2008. What's the point?: modal forms and attributes of Paleoindian bifaces in the New England-Maritimes region. Archaeol. East. N. Am. 36, 119–172. Burke, A.L., 1997. Lithic sourcing and prehistoric cultural geography in the Champlain Valley. J. Vt. Archaeol. 2, 43–52. Burke, A.L., 2007. Quarry source areas and the organization of stone tool technology: a view from Quebec. Archaeol. East. N. Am. 35, 63–80. Burke, A.L., Gauthier, G., Chapdelaine, C., 2014. Refining the Paleoindian lithic source network at Cliche-Rancourt using XRF. Archaeol. East. N. Am. 42, 101–128. Carty, F.M., Spiess, A.E., 1992. The neponset paleoindian site in Massachusetts. Archaeol. East. N. Am. 20, 19–37. Chapdelaine, C., 2012. The Early Paleoindian Occupation at the Cliche-Rancourt Site, Southeastern Quebec. In Late Pleistocene Archaeology & Ecology in the Far Northeast, edited by C. University Press, College Station, Chapdelaine general editor. Texas A&M. Condon, R.K., 1993. The structure and stratigraphy of South Mountain, west-central Vermont. reen Mt. Geol. 20 (1), 6.

7

Cox, S.L., 1998. Wired: investigating a high voltage Paleo-Indian site in Searsmont, Maine. Paper Presented at the North Easter Archaeological Association. Cox, S.L., Bourque, B., Corey, R., Doyle, R., Lewis, R., 1994. Phase II Investigations of Site 39.1, Searsmont, Maine. Crock, J.G., Robinson IV, F.W., 2012. Maritime mountaineers: Paleoindian settlement patterns on the west coast of New England. In: Chapdelaine, C. (Ed.), Late Pleistocene Archaeology and Ecology in the Far Northeast. Texas A&M Press College Station. Curran, M.L., 1984. The Whipple site and Paleoindian tool assemblage variation: a comparison of intrasite structuring. Archaeol. East. N. Am. 12, 5–40. Curran, M.L., 1994. New Hampshire Paleo-Indian research and the Whipple site. New Hampshire Archaeologist 33 (34), 29–52. Dillehay, T.D., 1989. Monte Verde: A Late Pleistocene Settlement in Chile, Vol. 1. Paleoenvironment and Site Context 1Smithsonian Institution Press, Washington D.C. Dillehay, T.D., 1999. The late Pleistocene cultures of South America. Evol. Anthropol. 7, 206–216. Dillehay, T.D., 2000. First Settlement of America: A New Prehistory. Basic Books, New York. Dincauze, D.F., 1996. Large Paleoindian sites in the northeast: pioneers' marshalling camps? Bull. Mass. Archaeol. Soc. 57 (1), 3–17. Donta, C., 2005. The Neponset Site, Locus 4: more evidence of a Michaud-Neponset phase occupation. Bull. Mass. Archaeol. Soc. 66 (2), 76–87. Ellis, C.J., 2011. Measuring Paleoindian range mobility and land-use in the Great Lakes/ Northeast. J. Anthropol. Archaeol. 30 (3), 385–401. Ellis, C., Lothrop, J.C., 1989. Eastern Paleoindian Lithic Resource Use. Westview Press, Boulder. Ferring, C. Reid 2001 The archaeology and paleoecology of the Aubrey Clovis site (41DN479) Denton County, Texas Gardner, W.M., 1989. An Examination of Culture Change in the Late Pleistocene and Early Holocene (circa 9200 to 6800 B.C.). In: Wyttkofski, J.M., Reinhardt, T. (Eds.), Paleoindian Research in Virginia: A Synthesis. Archaeological Society of Virginia Special Publication 19, Richmond, Virginia, pp. 5–52. Gauthier, G., Burke, A.L., Leclerc, M., 2012. Assessing XRF for the geochemical characterization of radiolarian chert artifacts from northeastern North America. J. Archaeol. Sci. 39 (7), 2436–2451. Gilbert, M.T.P., Jenkins, D.L., Gotherstrom, A., Naveran, N., Sanchez, J.J., Hofreiter, M., Thomsen, P.F., Binladen, J., Higham, T.F.G., Yohe II, R.M., Parr, R., Cummings, L.S., Willerslev, E., 2008. DNA from pre-Clovis human coprolites in Oregon, North America. Science 320 (5877), 786–789. Gilbert, M., Thomas, P., Jenkins, D.L., Higham, T.F.G., Rasmussen, M., Malmstrom, H., Svensson, E.M., Sanchez, J.J., Cummings, L.S., Yohe II, R.M., Hofreiter, M., Gotherstrom, A., Willerslev, E., 2009. Response to comment by Poinar et al. on “DNA from pre-Clovis human coprolites in Oregon, North America”. Science 325 (5937) (148-b-). Goodby, R.G., Bock, P., Bouras, E., Christopher, D., Garret Evans, A., Largy, T., Pollock, S., Rockwell, H., 2011. Phase III Data Recovery at the Tenant Swamp Site (27CH187), Keene Middle School Site, Maple Avenue, Keene. New Hampshire, Monadnock Archaeological Consulting, Stoddard, NH. Goodyear, A.C., 1979. A hypothesis for the use of cryptocrystalline raw materials among Paleo-Indian groups of North America. Research Manuscript Series No. 156. Institute of Archaeology and Anthropology, University of South Carolina, Columbia. Gramly, R.M., Rutledge, K., 1981. A new Paleo-Indian site in the State of Maine. Am. Antiq. 46 (2), 354–361. Gramly, R.M., 1982. The Vail site: a Paleo-Indian encampment in Maine. Bull. Buffalo Soc. Nat. Sci. 30. Gramly, R.M., 1984. Kill sites, killing ground, and fluted points at the Vail site. Archaeol. East. N. Am. 12, 101–121. Gramly, R.M., Storck, P.L., 1988. The Adkins Site: A Palaeo-Indian Habitation and Associated Stone Structure. Persimmon Press. Gramly, R.M., 1995. The Vail paleo-Indian site, Maine: fifteen years after the excavation. Amateur Archaeol. 1 (2), 46–64. Gramly, R.M., 2005a. Fluted drills and recycled fluted points at the Vail Clovis site. Amateur Archaeologist 11 (1), 47–71. Gramly, R.M., 2005b. The Morss fluted point site, northwestern Maine. Amateur Archaeologist 7 (2), 1–20. Gramly, R.M. (Ed.), 2009. Paleo-Americans & Palaeo-Environment at the Vail Site, Maine. Persimmon Press, North Andover, MA. Gramly, R.M., 2010. The Vail habitation and kill site: implications for Palaeo-American behavior and band size. Ohio Archaeologist 60 (3), 4–17. Hall, B., 1970. Stratigraphy of the Southern End of the Munsungun Anticlinorium. Maine, Maine Geologic Survey. Hazelwood, L., Steele, J., 2003. Colonizing New Landscapes: Archaeological Detectability of the First Phase In Colonization of Unfamiliar Landscapes. In: Rockman, M., Steele, J. (Eds.), The Archaeology of Adaptation. Routledge, London, pp. 203–221 general editor. Jones, G.T., Beck, C., Jones, E.E., Hughes, R.E., 2003. Lithic source use and paleoarchaic foraging territories in the Great Basin. Am. Antiq. 68 (1), 5–38. Kelly, R.L., Todd, L.C., 1988. Coming into the country: early Paleoindian hunting and mobility. Am. Antiq. 53 (2), 231–244. Kitchel, N.R., Barker, A., 2016. Results of a preliminary archaeological and geoarchaeological survey of the Munsungun Lake Formation, Pisquitaquis and Aroostook Counties, Maine. Submitted to the Maine Historic Preservation Commission. Copies Available from the Maine Historic Preservation Commission, Augusta. Meltzer, D.J., 1984. On stone procurement and settlement mobility in eastern fluted point groups. North American Archaeologist 6 (1), 1–24. Meltzer, D.J., 2002. What do you do when no one's been there before. Thoughts on the exploration and colonization of New Lands. In: Jablonsky, N. (Ed.), The First Americans:

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009

8

N.R. Kitchel / Journal of Archaeological Science: Reports xxx (2016) xxx–xxx

The Pleistocene Colonization of the New World. Memoirs of the California Academy of Sciences. University of California Press, pp. 25–56. Meltzer, D.J., 2009. First Peoples in a New World: Colonizing Ice Age America. University of California Press, Berkeley. Pollock, S.G., 1987. Chert formation in an Ordovician Volcanic Arc. J. Sediment. Res. 57 (1), 75. Pollock, S.G., Hamilton, N.D., Bonnichsen, R., 1999. Chert from the Munsungun Lake formation (Maine) in Paleoamerican archaeological sites in northeastern North America: recognition of its occurrence and distribution. J. Archaeol. Sci. 26, 269–293. Pollock, S.G., Hamilton, N.D., Boisvert, R.A., 2008. Archaeological geology of two flowbanded spherulitic rhyolites in New England, USA: their history, exploitation and criteria for recognition. J. Archaeol. Sci. 35, 688–703.. Prasciunas, M.M., T. Surovell 2015. Reevaluating the Duration of Clovis: The Problem of Non-Representative Radiocarbon Dates. In Clovis: on the Edge of a New Understanding, Texas A&M Press, College Station, edited by A. Smallwood and T. Jennings (general editor. Texas A&M Press, College Station). Rankin, D.W., Caldwell, D.W., 2010. A guide to the geology of Baxter State Park and Katahdin. Maine Geol. Survey, Bull. 43. Ratcliffe, N.M., Harwood, D.S., Hatch, N.L., Hepburn, J.C., Norton, S.A., Osberg, P.H., Stanley, R.S., 1975. Tectonic Studies of the Berkshire Massif, Western Massachusetts, Connecticut, and Vermont. United States Geological Survey Professional Paper:1. Reuther, J.D., Slobodina, N.S., Rasic, J.T., Cook, J.P., Speakman, R.J., 2011. Gaining Momentum: Late Pleistocene and Early Holocene Archaeological Obsidian Source Studies in Interior and Northern Beringia. In From the Yenisei to the Yukon: Interpreting Lithic Assemblage Variability in Late Pleistocene/Early Holocene Beringia, edited by T. Goebel and I. Buvit general editor. Texas A&M Press, College Station, TX. Ridge, J.C., Balco, G., Bayless, R.L., Beck, C.C., Carter, L.B., Dean, J.L., Voytek, E.B., Wei, J.H., 2012. The new north American varve chronology: a precise record of southeastern Laurentide ice sheet deglaciation and climate, 18.2–12.5 kyr BP, and correlations with Greenland ice core records. Am. J. Sci. 312 (7), 685–722. Robinson, B.S., Ort, J.C., Eldridge, W.A., Burke, A.L., Pelletier, B.G., 2009. Paleoindian aggregation and social context at Bull Brook. Am. Antiq. 74 (3), 423–477. Robinson, B.S., 2012. In: Chapdelaine, C. (Ed.), The Bull Brook Paleoindian Site and Jeffreys Ledge: A Gathering Place Near Caribou Island? In Late Pleistocene Archaeology & Ecology in the Far Northeast, pp. 182–190 (general editor. Texas A&M Press, College Station, TX). Rockman, M., 2003. Knowledge and learning in the archaeology of colonization. In: Rockman, M., Steele, J. (Eds.), Colonization of Unfamiliar Landscapes: The Archaeology of Adaptation. Routledge, London, pp. 3–24.

Sanchez, G., Holliday, V.T., Gaines, E.P., Arroyo-Cabrales, J., Martínez-Tagüeña, N., Kowler, A., Lange, T., Hodgins, G.W.L., Mentzer, S.M., Sanchez-Morales, I., 2014. Human (Clovis)–gomphothere (Cuvieronius sp.) association ∼13,390 calibrated yBP in Sonora, Mexico. Proc. Natl. Acad. Sci. 111 (30), 10972–10977. Smith, G.M., 2010. Footprints across the blackrock: temporal variability in prehistoric foraging territories. Am. Antiq. 75 (4), 865–885. Snow, D., 1980. The Archaeology of New England. Academic Press, New York. Spiess, Arthur, Wilson, D., 1987. Michaud: A Paleoindian Site in the New EnglandMaritimes Region. Maine Historic Press, Portland. Spiess, A.E., Wilson, D., Bradley, J., 1998. Paleoindian occupation in the New England-Maritimes region: beyond cultural ecology. Archaeol. East. N. Am. 26, 201–264. Tankersley, K.B., 1989. A close look at the big picture: early Paleoindian lithic resource procurement in the midwestern United States. In: Ellis, C.J., Lothrop, J.C. (Eds.), Eastern Paleoindian Lithic Resource UseInvestigations in American Archaeology. Westview Press, Boulder. Tankersley, K.B., 1991. A geoarchaeological investigation of distribution and exchange in the raw material economies of Clovis groups in eastern North America. Raw Material Economies among Prehistoric Hunter-Gatherers. 19, pp. 285–303. Waters, M.R., Stafford Jr., T.W., 2007. Redefining the age of Clovis: implications for the peopling of the Americas. Science 315 (5815), 1122–1126. Waters, M.R., Forman, S.L., Jennings, T.A., Nordt, L.C., Driese, S.G., Feinberg, J.M., Keene, J.L., Halligan, J., Lindquist, A., Pierson, J., Hallmark, C.T., Collins, M.B., Wiederhold, J.E., 2011a. The Buttermilk Creek complex and the origins of Clovis at the Debra L. Friedkin site, Texas. Science 331 (6024), 1599–1603. Waters, M.R., Stafford, T.W., Gregory McDonald, H., Gustafson, C., Rasmussen, M., Cappellini, E., Olsen, J.V., Szklarczyk, D., Jensen, L.J., Thomas, M., Gilbert, P., Willerslev, E., 2011b. Pre-Clovis mastodon hunting 13,800 years ago at the Manis site, Washington. Science 334 (6054), 351–353. Wilson, D., 2003. The Nicholas Giuseppe Colombo Lautman Site report. On file at the Maine Historic Preservation Commission. Witthoft, J., 1952. A Paleo-Indian site in eastern Pennsylvania: an early hunting culture. Proc. Am. Philos. Soc. 96 (4), 464–495. Wyckoff, D.G., 1993. Gravel sources of knappable Alibates silicified dolomite. Geoarchaeology 8 (1), 35–58.

Please cite this article as: Kitchel, N.R., Questioning the visibility of the landscape learning process during the Paleoindian colonization of northeastern North America, Journal of Archaeological Science: Reports (2016), http://dx.doi.org/10.1016/j.jasrep.2016.10.009