Property Types

.
Contents

 Introduction 

 Geography and  Environmental Conditions  

 Chronological  Considerations 

 Initial Human Occupation 

 Widespread Settlement 

 Terminal Paleoindian  Occupations 

 Initial Holocene Early  Archaic Assemblages 

 Property Types 

 Resource Distribution 

 Research Needs and  Questions 

 Evaluation Criteria 

 Possible NHLs in the  Southeast 

 Conclusions 

 Annotated References 

 References Cited 

 

Each of the major property type categories presented here is defined in the national context document. In the following paragraphs each property type is discussed in turn. Examples of specific sites or districts that illustrate the category in the Southeast are presented in the resource distribution section that follows. A listing of southeastern sites and localities by property type, and NRHP and NHL status, is provided in Table 3. Because there is considerable overlap between property types, some sites are listed and subsequently discussed under more than one category.

Isolated Finds
Isolated finds are individual artifacts that have been demonstrated to be Paleoindian in age through typological or other analyses. In the Southeast isolated finds tend to be projectile points that either occur by themselves or are found on sites with materials whose age is known or assumed to be later in time. Several thousand isolated finds of Paleoindian points and other tools are currently known from the region (Anderson and Faught 1998, 2000). Typically the isolated points appear to reflect individual hunting/butchering episodes, or examples of artifact loss or discard. The distribution of isolated finds, when examined collectively over large areas, has done a great deal to improve our understanding of Paleoindian settlement in the Southeast (Anderson 1990a, 1990b, 1996; Williams and Stoltman 1965). It is crucial to determine, of course, whether isolated finds are parts of larger, unrecognized sites, as an important cautionary tale illustrates. In the Tennessee fluted point survey, on more than one occasion, the "isolated finds" of several individual collectors were found, upon cross checking, to come from the same location, and hence derive from significant sites (Broster et al. 1996:9). Likewise, as Tesar (1996:22) has noted, "the collection and reuse of older artifacts should always be considered when they occur as minority elements in otherwise later assemblages." That is, isolated or small numbers of Paleoindian artifacts on sites with later components may represent scavenging and re-use rather than a primary occupation. Tesar (1994:7) has also argued that the replication of earlier tool forms by later occupants should also be considered when examining early materials.

Caches
Caches are groups of artifacts or other resources intentionally left at a location for either ceremonial or utilitarian purposes. Paleoindian caches tend to fall into two types, elaborate finished tools presumably associated with burials or consumption-rituals of some kind, and hence permanently removed from the cultural system of which they were a part, or mundane tools, raw materials, or foodstuffs left at a particular place with the intent of being used at a later date. Caches may be subsets of other property types if, for example, they occur on occupation sites, or as burial furniture.

No Paleoindian ceremonial caches predating roughly 10,800 rcbp/12,900 B.P. are known from the southeastern United States. Given the striking similarities of Clovis assemblages over much of the continent, and the presence of elaborate Clovis caches from a number of locations in the west, such caches probably do exist in the region. Ceremonial caches are well known from the Dalton culture of the central Mississippi Alluvial Valley (Morse, ed., 1997; Walthall and Koldehoff 1998). Well published examples include Sloan and the Hawkins Cache from northeast Arkansas (Morse 1971a, 1975a; Morse, ed., 1997).

Intact bifacial core/preforms or large unifacial flakes have been found at a number of southeastern Paleoindian sites, both at and away from quarry/reduction areas, and some of these artifacts are thought to be utilitarian raw material caches. Examples of these kind of artifacts have been reported in Hardaway/Dalton context at the Haw River sites in North Carolina (Cable 1982b:321), in Suwannee/Early Side Notched context at Harney Flats (8Hi507) in Florida (Daniel and Wisenbaker 1987:81-83), in Clovis context at the Conover site in southern Virginia (McAvoy 1992:108, 115), and in unknown but presumably Clovis context at the Big Pine Tree (38AL143) and Smith's Lake Creek (38AL135) sites at the Allendale chert quarries in South Carolina (Goodyear 1992, 1999a:458-462; Goodyear and Charles 1984; Goodyear et al. 1985), and at Wakulla Springs Lodge (8Wa329) in Florida (Jones and Tesar 2000; Tesar 2000).

In the Southeast, large blades and blade cores have been found with Clovis and related fluted point assemblages at the Carson-Conn-Short site in Tennessee (Broster and Norton 1996:290-293; Broster et al. 1994, 1996; Nami et al. 1996) and the Adams site in Kentucky (Sanders 1988, 1990). The large size of some of the specimens found at these sites suggest they may have been cached, although profligate use of readily available raw material has also been suggested as an explanation. The terminal Paleoindian levels at Dust Cave, Alabama, dating to after ca. 10,500 rcbp/12,500 B.P., contained a large number of scraping tools made on blades, some of which may have been cached in the rockshelter (Meeks 1994). Although just outside the Southeast as defined for purposes of this study, at Rodgers Shelter in Missouri river cobbles and the bones of a trumpeter swan were found in Dalton levels, and have been interpreted as raw material caches for stone and bone working (Walthall 1998:229). The Busse cache in Kansas, of presumed Clovis age, evinces both ceremonial and utilitarian functions, consisting as it does of a large quantity of lightly to heavily utilized chert bifaces, blades, and other flaked stone tools, some of which are streaked with red ochre (Hofman 1995; see also Roper 1996).

While no unequivocal evidence for pit features has been found on Paleoindian sites in the Southeast, such features, if ultimately shown to be present at some sites, could have been used to store nuts, as they were in later times. Some southeastern Paleoindian sites, particularly those found associated with springs, streams, or ponds, may have been subaqueous meat caches rather than strictly kill sites (cf., Fisher 1995).

Bone Beds and Kill Sites
No unambiguous pre-Clovis Paleoindian kill sites or bone beds are known from the Southeast, although a number of possible candidate sites exist, including Saltville in Virginia, Little Salt Springs, Page Ladson, and Sloth Hole in Florida, and Coats-Hines in Tennessee (Breitburg et al. 1996; Clausen et al. 1979; Dunbar et al. 1988; Hemmings 1998; McDonald 2000). A number of Clovis age sites exist where evidence for the killing or butchering of fauna has been found, and modified bone and ivory of extinct Pleistocene fauna have been found at a number of locations in Florida (some of which may be pre-Clovis in age) (Dunbar and Webb 1996; Dunbar et al. 1989; Webb et al. 1984). Remains of essentially modern Holocene fauna have been found at a number of later Paleoindian sites, typically with Dalton occupations. These are usually from rockshelters, locations that offer unusual preservation, and that were first intensively utilized during this period (Goldman-Finn and Walker 1994; Styles and Klippel 1996; Walker 1997, 2000; Walthall 1998).

While not true kill sites, or even dense bone beds, Paleoindian sites with well preserved faunal and floral remains are of crucial importance to understanding human adaptation during this period. Well preserved later Paleoindian age faunal remains have been found in a number of rockshelters in the Southeast, including at Dust Cave and the Stanfield-Worley bluff shelter in Alabama (Goldman-Finn and Walker 1994; Parmalee 1962; Walker 1997, 2000). The faunal assemblages from these sites encompass a wide range of species, from a variety of environments. Of particular importance is the fact that appreciable numbers of small mammals, reptiles, fish, and birds are represented in these samples. At Dust Cave, in fact, birds are extremely common in the later Paleoindian Dalton assemblage (Walker 2000). Dincauze and Jacobson (2000) and Fiedel (n.d.) have each suggested that birds were important in Paleoindian diet, and that following migratory birds may have prompted and facilitated group movement and even colonizing behavior. White-tailed deer have traditionally been assumed to have been of great economic importance to Dalton populations, and the Dalton toolkit, in fact, was at one time interpreted largely in terms of its utility for the bulk processing of deer meat and hides (Goodyear 1974:14; Morse 1973). We now believe that these populations had a highly diversified subsistence economy, in which small game played an important if not primary role (Walker 1997, 2000).

It would be important to document how important small game was to earlier Paleoindian populations, for which few sites with paleosubsistence remains other than megafauna have been found. While it has sometimes been suggested that Clovis points were used exclusively on large game animals, it appears this may not have been the case. Artifacts such as stone tools may be examined to provide direct evidence about the range of animal and plant species exploited. Blood residue immunological analyses undertaken to date on materials from the Cactus Hill, Fifty (44WR50), and Thunderbird (44WR11) sites in Virginia suggest a wide range of species may have been exploited (Newman 1994, 1995). Positive results to deer and elk antiserum were identified on one fluted point from Cactus Hill, to cat antiserum on a fluted point from Thunderbird, and to rabbit and bear from separate fluted points from Fifty. Positive reactions for deer, elk, rabbit, and an unidentified bovid, possibly bison or muskox, were reported from several unifacial tools from Cactus Hill (reported in McAvoy and McAvoy 1997:Appendices F and G). While still somewhat controversial, blood residue analysis can provide direct evidence about the kinds of animals exploited by these early populations.

The limitations of blood residue analysis, however, mean these results will need to be considered carefully and, where possible, subject to additional verification through blind testing and independent replication (Eisele et al. 1996; Fiedel 1996). That is, care in the collection of artifacts in the field and in subsequent handling is essential if blood residue analyses are to be undertaken, since contamination is a distinct possibility, and indeed the method itself may be unreliable (cf., Loy and Dixon 1998, with Downs and Lowenstein 1995; Eisele et al. 1995, and Fiedel 1996). Phytolith analysis, a method of resolving patterns of plant use, has yet to be conducted on southeastern Paleoindian stone tools. Phytolith analysis has, however, been used to explore site formation processes at the Cactus Hill site, in a recent innovative analysis by McWeeney (McAvoy et al. 2000). Phytolith incidence was found to covary with cultural remains and soil phosphate content, indicating these remains may be useful in delimiting occupation floors.

Other kinds of special use or extraction sites, such as fish weirs, fishing or aquatic mammal hunting stations, or shellfish gathering sites, may also be present in coastal areas, although these would have likely been eroded and submerged by rising sea-levels (Tesar 1996:30-31). Submerged sites have been found at appreciable distances out from the Gulf and Atlantic coasts in recent years (Blanton 1996; Faught 1996; Faught and Carter 1998; Faught and Donoghue 1997; Faught et al. 1992). While shell midden sites dating prior to the Mid-Holocene in age are currently unknown in the region, their existence is considered possible (Russo 1996:196).

Human Burials
Human burials of Late Pleistocene age are rare in the Southeast, as indeed they are throughout North America. The only well documented, unequivocal burial assemblage of Paleoindian age known from the region comes from the Sloan site in northeast Arkansas, where some 200 tiny, weathered bone fragments were found amid a remarkable assemblage of Dalton points and other tools (Condon and Rose 1997; Morse 1975a, Morse, ed., 1997). The Sloan case indicates that human burials of Late Pleistocene and Early Holocene age may occur in unusual or unanticipated settings. Sloan was located well away from contemporaneous Dalton sites, for example, on a sand dune that saw only minor use in later prehistory. Although there are many other Dalton sites known from this part of northeast Arkansas, most are assumed to have been either habitation or special purpose resource extraction stations. Sloan, in contrast, appears to have been used only for mortuary behavior.

Large numbers of Archaic period burials dating from ca. 8300-5200 rcbp/9350-5930 B.P. have been found in submerged settings in Florida in recent years, at sites like Little Salt Springs and Windover (Clausen et al. 1979; Doran et al. 1986). The earliest submerged human remains from the region at present have been dated from ca. 9000 to 10,000 rcbp/10,200 to 11,450 B.P., at the Warm Mineral Springs site in Florida (Clausen et al. 1975a, 1975b; Cockrell and Murphy 1978; Goodyear 1999a:445; Tesar 2000). A total of 33 dates were obtained from the zone where human remains were found, a ledge ca. 12 to 14 m below the surface, which average ca. 9630 rcbp/11,000 B.P. (Tesar 2000:12). Associated artifacts included Greenbrier/Bolen projectile points, suggesting a late Dalton/Early Side Notched cultural assemblage. Tesar (2000:13-14, citing arguments presented by Clausen et al 1975a:31) has argued that, rather than intentional burials, the human remains are those of drowning victims who accidentally fell or deliberately climbed into the cenote but were subsequently unable to climb out. Recent investigations have focused on the central debris cone at the bottom of the sinkhole, in more than 150 feet of water, where relatively undisturbed deposits may exist (Purdy 1991:187-188; Tesar 2000:14). In the absence of a formal site report, interpretation of findings is difficult. Numerous burials (possibly >1000) have also been found at the nearby Little Salt Spring site, which yielded possible pre-Clovis artifacts, but the burials are Mid-Holocene in age, from ca. 6800 to 5200 rcbp/7650 to 5930 B.P. (Clausen et al. 1979:612; Tesar 2000:17-20).

The Page-Ladson site in Florida has yielded well preserved floral and fauna remains (Peres 1997; Peres and Carter 1999), and human remains may eventually be discovered as well. Early human remains may even be present in offshore contexts. A partially mineralized human bone fragment was found washed up on Edisto Beach, South Carolina, possibly derived from offshore Pleistocene deposits, and dated to 6960±240 rcbp/7775 B.P. (Hemmings et al. 1973). Paleoindian artifacts have been found at appreciable distances out onto the continental shelf, and skeletal materials may also be present (Faught 1996; Goodyear 1999a:468-470). These sites are amenable to examination using underwater archaeological techniques, a type of research that shows great promise.

Other Paleoindian burials no doubt exist within the region, although when discovered, their context will need to be carefully evaluated. The Natchez pelvis, for example, was found in indirect association with Late Pleistocene fauna, but was radiocarbon dated to much later in time (Cotter 1991). Likewise, some human skeletal specimens found and dismissed long ago might bear re-examination, such as the human remains from Vero and Melbourne in Florida (Gidley and Loomis 1926; Sellards 1917).

Rock Art and other Petroglyphic or Pictographic Representations
No examples of Late Pleistocene age rock art or other stationery artwork are currently known from the Southeast, although their existence cannot be ruled out. Most known fixed artwork in the region, in fact, is quite late, dating from the Woodland or Mississippian periods, as exemplified by sites like Mud Glyph Cave in Tennessee (Faulkner 1986; Simak et al. 1997). Designs drawn into mud deposits in caves, in fact, appear at present to be more common than petroglyphs or pictographs in the Southeast. In the western and southwestern parts of the country, in contrast, petroglyphs and pictographs are quite common, and some appear to have great antiquity (Schaafsma 1996:599-600). This may reflect local cultural traditions as well as factors influencing preservation. Outside of isolated and climatically stable cave environments, the long term survival of rock art may be unlikely in the humid southeastern climate. Recent rock art surveys in southern states, however, indicate the record is more extensive than once thought, suggesting people may not have been looking in the right places, or carefully enough (e.g., Charles 1998).

Parietal art, "the art restricted to the walls, roof and occasionally floors of caves and rock-shelters" (Clark 1967:67), can also only occur in areas where caves and rockshelters themselves are likely. Fortunately, geological formations conducive to cave and rockshelter formation occur over large parts of the Southeast, particularly in the interior highlands and plateaus. Only in the Coastal Plain are such settings rare, although a major exception is the karst topography in Florida. In this region, many of the caves and shelters that would have been exposed in the Late Pleistocene are now submerged, with unknown effects on the survival of possible artwork, as well as making the exploration of the question difficult. Evidence for Late Pleistocene age painting has been observed on the walls and ceiling of the Pedra Pintada rockshelter near Monte Alegre in Brazil, with painted fragments found in strata securely dated to ca. 10,500 rcbp/12,500 B.P. (Roosevelt et al. 1996). Given proper preservational conditions, artwork of a similar age may eventually be found in the Southeast. Surveys of rock art have been initiated in some southeastern states, such as South Carolina (Charles 1998), and more information about the nature and age of these kind of remains should be forthcoming.

Chattel art, or arte mobilière, consisting of small, portable items of carved or decorated wood, bone, and stone, may also be discovered some day in Late Pleistocene context in the Southeast. Indeed, the elaborately carved ivory and bone points and foreshafts found in Florida are considered by some to be works of art as well as utilitarian items. Similar aesthetic qualities are attached to unusually large or well made flaked stone tools, such as Sloan points, or some of the more spectacular Clovis and Cumberland points found in the region.

Quarries and Workshops
Quarries and workshops comprise perhaps the best known and certainly among the most easily recognized Paleoindian property type in the Southeast. Major sites and localities include Theriault in Georgia (Brockington 1971), the Little River District in Kentucky (Sanders 1990; Freeman et al. 1996), Hardaway in North Carolina (Coe 1964; Daniel 1998), the Allendale quarries in South Carolina (Goodyear and Charles 1984), Carson-Conn-Short (Broster et al. 1994) and Wells Creek Crater (Dragoo 1973) in Tennessee, and the Flint Run complex (Gardner, ed., 1974) and the Williamson site (McCary 1951, 1975) in Virginia (Table 3). At these sites, lithic raw materials were extracted and initially processed for use at other locations. Many quarry/workshop sites are also multicomponent, with evidence for Clovis occupations, as well as later Paleoindian through Mississippian period use. Occupation or habitation by Paleoindian groups is explicitly stated to have occurred at almost all of these sites (e.g., Daniel 1998:145; Dragoo 1973:46; Gardner 1977:258-259; McAvoy 1992:142; Sanders 1990:62). Accordingly, strict separation of quarry/workshop, and occupation property types is difficult or impossible in some cases, as is the resolution of discrete Paleoindian assemblages from those of much later periods. Researchers examining these property types should make every effort to resolve intrasite spatial patterning, which may vary appreciably over relatively small areas.

Occupations
A great many Paleoindian occupation sites, defined as habitation areas or residential base camps occupied for unknown but presumably fairly lengthy periods of time, are known from the Southeast (Table 3). Many quarry/workshop sites also appear to have occupation areas within them or nearby, such as at Flint Run in Virginia, where a possible structure has been identified (Gardner 1974), or Hardaway in North Carolina, where highly diversified tool assemblages have been found (e.g., Coe 1964; Daniel 1998). There are also a number of presumed Paleoindian habitation sites in the Southeast that are not located in direct proximity to lithic raw material sources, yet that have been extensively examined and that are well published. These include Dust Cave (Driskell 1996; Goldman-Finn and Driskell 1994) and Stanfield-Worley in Alabama (DeJarnette et al. 1962), Brand in Arkansas (Goodyear 1974), the Haw River sites in North Carolina (Claggett and Cable 1982), and Taylor in South Carolina (Michie 1996), to cite a few examples that provide appreciable insight into Paleoindian lifeways. The assemblages at Paleoindian occupation sites can also be used to derive expectations of what habitation assemblages may be like within quarry/workshop sites, where lithic raw material quarrying and initial reduction occurred, and where the associated massive debris likely masks their easy recognition.

Return to the Top