Florida Historical Contexts: The Paleoindian Period

The Paleoindian Historical Context was written by Nina Borremans as part of a project directed by Gerald T. Milanich in 1990. Illustrations were added later and some text sections have been updated by staff of DHR.

Introduction

Archaeological evidence indicates that, in addition to hunting small game and gathering wild plants and shellfish, Paleoindians stalked now-extinct Pleistocene megafauna; however, the role these large animals played in the day-to-day business of making a living varied considerably across the continent. Large, stemless, stone spear points, the most diagnostic of the artifacts made by Paleoindians, are time-markers used by archaeologists to identify Paleoindian sites. Due to the ravages of time, erosion, and decomposition of perishable artifacts, projectile points and other stone artifacts are often all that remain to represent these early prehistoric people and their hunting-gathering way of life.

The environmental repercussions of climatic change and glacial melting provided a dynamic background for cultural response during the Paleoindian period, which lasted about five thousand years. The end of the Paleoindian way of life is difficult to identify archaeologically and may have been one of gradual transition.

About 10,000 years ago, changes in the Florida landscape and environment coincided with new developments in subsistence technology and settlement patterns, leading to the next archaeologically defined stage, the Archaic.

Less than 100 Paleoindian sites are recorded in Florida. These known sites are scattered around the state and their recording is primarily a result of accidental discoveries. Many more Paleoindian sites undoubtedly exist, but they are located offshore on the continental shelf, in terrestrial wet areas, or are deeply buried. These inaccessible locations make it difficult to identify Paleoindian sites, and our ignorance of them has biased our interpretation of Paleoindian culture.

The Setting

The Paleoindians lived in a Florida twice the size it is today. At the time they lived, sea level was 60-100 m lower, exposing vast expanses of the present continental shelf (Gagliano 1977; Blackwelder et al. 1979). Present-day coasts were inland, even upland, areas. The late Pleistocene shorelines in the Gulf of Mexico were located as much as 120 to 150 km seaward of their present locations. It is not difficult to see why Paleoindian period coastal sites have yet to be discovered in Florida-they are submerged beneath scores of fathoms of ocean water, tens of kilometers offshore (Stright 1986; Garrison 1989).

Pollen and paleontological studies have provided us with evidence of the climate and environment at this time. They indicate that Florida was considerably drier than it is today. Vegetation of north Florida highlands at about 14,000 B.P. generally was open pine forests giving way to oak/hickory stands and local prairies. The central and southern peninsula had open xeric scrub vegetation. Climatic condition of the coasts is still a subject for debate among paleoecologists. The most common view is that the coastal strips remained arid throughout the Paleoindian period, supporting savannahs and dune scrub except where springs and rivers brought life to the dry conditions.

Paleoecological reconstructions focusing on macrobotanical and paleontological samples give us a picture painted with a finer brush. They suggest that the environmental mosaic was more diverse than the pollen studies can indicate. Mesic vegetation punctuated the forests and savannahs. Seepage and runoff from springs and rivers provided water for moisture-loving trees and shrubs, which were limited elsewhere by lack of groundwater.

These hammocks supported a suite of animal and plant life that exceeded what would have been available in the dry forests alone. By 10,000 B.P., the forests became denser, while oaks and pines filled in previously unforested areas. Oak savannahs replaced much of the scrub vegetation of the lower peninsula.

Material Culture

Of all of the materials used by the Paleoindians for clothing, shelter, tools, weapons, ornaments, containers, etc., only stone has stood the test of time. Artifacts made of organic materials such as skin, shell, bone, wood, and plant fiber have long since perished in the acid Florida soil, with the exception of those that have been preserved in submerged deposits. As a result, our knowledge of the Paleoindian artifact assemblage is largely limited to lithic tools and the by-products of their manufacture and use. The Paleoindians had a marked preference for high quality chert (silica-replaced limestone) and fine workmanship.

Paleoindian research has, until recently, focused on the definition and description of projectile points. As more sites are found with intact stratification, however, investigators will be able to give more attention to the lithic complex as a whole. By studying chipped stone remains, archaeologists learn how chert was quarried from limestone outcrops, how the raw material was modified to produce tools, and to what uses these artifacts were put. Of equal interest is how each of these activities and processes may have varied from site to site and regionally in Florida, and how they changed through time.

Large, lanceolate projectile points have been recognized by archaeologists as the hallmark of the Paleoindian period. Some of these points may have been used as hafted knives. The Suwannee point is the most commonly reported lanceolate in Florida, but several other types and varieties have been defined to incorporate the stylistic and temporal variations in form. Common traits include lateral rather than basal thinning, basal grinding, and straight to slightly waisted lower sides. Based on technological and stratigraphic investigations, some temporal trends in projectile point shape have been documented. Point length and thickness appear to decrease in time, while waisting increases.

Analyses of Paleoindian artifact assemblages show that most tools were generalized in nature. They appear to be multi-functional and do not represent the specialized forms one might expect if only certain kinds of animals were hunted and processed to the exclusion of others (Daniel and Wisenbaker 1987).

A number of unifacial lithic tools have come to be considered formal Paleoindian artifact types because they display regular morphological features, reflect substantial effort in manufacture, and can be recognized in assemblages from other Paleoindian sites in Florida and the rest of the Eastern United States. Typically, these tools are plano-convex, exhibit steeply flaked working edges, and appear to have been used primarily for scraping.

Paleoindian unifacial scrapers vary widely in shape and size, presumably reflecting the wide variety of tasks for which they were employed. By analyzing length measurements and length:width/length:height ratios, a rough typology has been developed. The Paleoindian artifact assemblage from the Harney Flats site (8HI507), probably the most complete assemblage known from Florida, has been described in detail by Daniel and Wisenbaker (1987). Among the categories of unifacial tools they discuss are endscrapers (including thumbnail scrapers) and discoidal scrapers (Daniel and Wisenbaker 1987:65-74).

Also found in the Paleoindian tool complex is an array of tools that do not exhibit diagnostic traits. The majority show only slight modification and use wear. These artifacts are often interpreted as informal and expedient tools. Blade knives, flake knives, gravers, and bifacial knives are recovered. Small retouched flakes, flake debitage, cores, and hammerstones are also frequently found at intact Paleoindian sites.

Few of the artifacts found at Paleoindian sites are made of material other than chert. Sandstone hones or abraders are probably part of the tool kit. Egg-shaped objects made of ground stone may have been used as bola weights in the hunting of water birds and other game (Milanich and Fairbanks 1980:39).

These interpretations suggest that fishing may have been of greater economic importance than current subsistence models allow. Rarer than the bone points and foreshafts are other bone tools, socketed antler projectile points, worked shell, and worked fossil shark's teeth.      

Bone and ivory foreshafts and bone pins recovered from springs, lakes, and rivers constitute the majority of artifacts made of organic materials. The two-piece foreshafts were used in hafting projectile points to wooden shafts. Double-pointed bone implements are commonly found in rivers near kill sites. Hypotheses concerning their use range from leisters or harpoon tips to awls to fish hooks.

Plant materials must have played an important role in Paleoindian technology because they do in every other sub-Arctic culture. Underwater excavations at Little Salt Spring (8SO18) resulted in the recovery of several wooden artifacts, including a stake used to impale a now-extinct species of giant land tortoise, a carved oak log mortar (for grinding seeds or nuts), and a nonreturning boomerang made of oak (Clausen et al. 1979). Recent excavations at the submerged Page-Ladson site (8JE591) on the Aucilla River have resulted in the recovery of preserved plant remains that are a valuable source of paleoecological information (Dunbar et al. 1988). Continued searching for Paleoindian wet sites may someday lead to the discovery of a complete archaeological assemblage representing the breadth of techniques and materials employed.

Subsistence

The Paleoindians have conventionally been characterized as hunting and gathering nomads, moving seasonally as the availability of game and wild plant foods changed, settling only for brief periods when resources such as nuts or fruits were temporarily plentiful. Although most reconstructions of Paleoindian subsistence have emphasized the role of large now-extinct Pleistocene animals like mastodons, many researchers now suggest that the Paleoindian diet was more generalized and included smaller game, fish, shellfish, and plant foods. We now know from studies by Daniel and Wisenbaker (1987) and others that Paleoindians were not as nomadic as previously believed.

Our perceptions about the Paleoindian lifestyle in Florida have been colored by the emphasis on the Paleoindian Big Game Hunting Tradition described for the Western United States. We now think that their diet included much more than elephants and bison on a day-to-day basis, although faunal and botanical remains are rarely recovered in direct association with Paleoindian artifacts in residential sites (spring and sinkhole sites constitute an important exception).

Models of Paleoindian lifeways in Florida should consider physiography, climate, vegetation, and animal populations in assessing the potentials for food and raw material resources.

In addition, these assessments should be undertaken on regional and local scales. In Florida, late Pleistocene savannahs supported grazing herds somewhat analogous to those found on the Plains, but many other animals (large and small) lived in the hammocks surrounding the rivers, lakes, and sinkholes that were important locations for Paleoindian activities (Webb and Martin 1974).

Evidence of extensive interaction with other groups can be seen in the uniformities observed in artifact types across the continent. Projectile points of this period show only minor regional variations in form. Occasional pieces of exotic materials are found at Paleoindian sites in Florida, suggesting interregional travel and trade. However, most utilitarian artifacts found in areas where chert is accessible were made of materials found nearby, suggesting that the lifestyle of these early Floridians may not have been quite as nomadic as we have supposed (see Goodyear et al. 1983).

Major factors in questions of sedentism in Paleoindian settlements are the reliability and richness of the resource base. Although human population density is thought to have been very low, putting minimal strain on prey populations, the seasonal vagaries of terrestrial resource availability would have been a major factor affecting sedentism. Aquatic ecosystems, on the other hand, have been shown to provide a relatively stable, nutrient-rich food base when compared with terrestrial habitats. The role of aquatic resources is rarely included in discussions of the Paleoindian lifeway despite the fact that sites of this age are most frequently found in or near water. Freshwater fish, turtles, alligators, and shellfish may have been as important to the Paleoindians as they were for their successors, but a lack of preserved faunal remains prevents an evaluation of this thesis. The hypothesis that Paleoindians developed or brought with them a tradition of maritime adaptation is also difficult to document (or refute) at this time. Both the Atlantic and the Gulf coasts of Florida would have provided fish and, perhaps, shellfish, and it is unlikely that such potentially important resources were ignored. As our means of underwater site detection and testing improve, we may well find 10,000 year old shell middens 120 km offshore.

Settlement Patterns

The predictive site location model developed by Dunbar and Waller (1983) focuses on the close correlation between known Paleoindian sites and the distribution of chert-bearing limestone outcrops. While the availability of chert for tool production was certainly a factor in site location, the site clusters identified are also correlated with surface water availability. With a dry climate and sea level 60-100 m below its present stand, inland water tables were also low, (recognizing that there may not be a direct correlation between coastal sea level elevations and interior local groundwater levels) resulting in a scarcity of available water.

Most of the sites within the mapped distribution are located in or near sources of permanent fresh surface water, which must have been a valuable resource for people and game animals alike.

Chert is most accessible in many of the same locations where sites have been found but also where ancient sites have the best chance of being discovered. These include: 1) where erosion has removed an overlying mantle of clastic materials, such as along inland waterways, 2) where the rate of sediment accumulation is low, such as on the tops and slopes of hills, 3) where karst features puncture the sedimentary overburden, such as sinkholes, and 4) on the central Gulf coast, where erosion due to sea level rise has bared limestone bedrock. Paleoindian sites have a higher probability of being found in these situations than in other areas where thick deposits of sands have accumulated.

As demonstrated at Harney Flats, testing in areas where Paleoindian artifacts have been found can locate sites even if they are buried under thick sand deposits. To date such testing has been limited in Florida.

Modern alterations of the Florida landscape to increase the amount of dry land available for buildings, farms, and roads have resulted in increased access to early prehistoric sites. Many sites would not be discovered using accepted survey techniques due to their location in former swamps. This points out the high probability that most Paleoindian sites will be missed by conventional surveys since most sampling strategies target only land that is well to moderately drained.

If most early prehistoric sites are deeply buried, surveys are unlikely to identify Paleoindian sites because shovel tests are generally too few, too small, and too shallow to intersect them. Unless exposed by bulldozing, road cuts, or natural erosion, most buried Paleoindian sites will be found because they happen to underlie younger, more accessible sites. Clearly, we must revise our testing strategies.

The lanceolate projectile points characteristic of this time period are uncommon along the coasts of Florida with the exception of the north central Gulf coast where Tertiary karstic terrain brings both chert and water to the surface (Dunbar and Waller 1983). Quite probably, many terrestrial Paleoindian sites remain undiscovered along the coasts because they occur below the current water table and have been covered by soils in these lowlying areas.

Some archaeologists are giving increasing attention to the high probability of locating drowned Paleoindian and Archaic sites on continental shelves worldwide. Geologic studies of drowned riverine, lagoon, and marsh deposits indicate that estuarine resources may have been continuously available from Paleoindian through Archaic times. Recent efforts devoted to the discovery of coastal sites have met with some success using a site location model focussing on drowned rivers and submerged limestone outcrops. Talking with fishermen and sport divers has resulted in the identification of submerged shell midden sites (not Paleoindian) in the Tampa Bay area and in the drowned channel of the Aucilla River (Ruppé 1980; Dunbar 1989). Diagnostic projectile points dredged from Tampa Bay and the Atlantic coast provide evidence that submerged Paleoindian sites exist, although the search for maritime sites must look farther offshore (Goodyear et al. 1983).

Remote sensing techniques employed in the Gulf of Mexico have identified Pleistocene sinkholes on the outer continental shelf, attesting to the potential for locating deeply submerged Paleoindian sites in the future in this manner (Garrison 1989). Archaeologists may eventually identify potential site locations with side scanning sonar, sub-bottom profiling, and a host of other sophisticated techniques, then sample them using submersibles.

Base Camps and Villages

Quarries

Quarries are locations where raw material (chert) was mined and processed for tool manufacture, and they occur in the vicinity of outcrops at or near the ground surface. In addition to large chunks and nodules of chert, cores, flakes, and other evidence of first stage manufacturing are generally recovered. When outcrops occur near a water source, a residence site is often found nearby. Quarry sites are common in north central Florida (especially Marion and Alachua counties) (Purdy 1975, 1981), in the central Gulf coast area (especially Hillsborough County), in north Florida along the Santa Fe, Suwannee, and Aucilla Rivers, and in northwest Florida along the Chipola River.

Short Term Camps

Kill Sites

Paleoindian artifacts have been found in association with animal bones in a number of locations in rivers, sinkholes, and spring runs (e.g., the Guest Mammoth [8MR130] and Little Salt Spring [8SO18] sites). Animals were killed and/or butchered at these sites. The artifact inventory is similar to the short term camp including projectile points, utilized flakes for cutting and scraping, and waste flakes from use or resharpening. Most of the river finds are thought to have come from kill sites; however, because they are found in deflated, mixed contexts the association of the bones and artifacts is often in question.

Isolated Projectile Points Most of the diagnostic Paleoindian artifacts are recovered without associated materials. While they provide little in the way of information, they are useful indicators of human geography and technological diversity. Isolated projectile points are often found in the rivers and sinkholes of north central Florida, and on the surfaces of river banks and hilltops. Projectile points are also occasionally dredged from marine channels or found on sand bars at low tide by fishermen in the Gulf of Mexico (Goodyear et al. 1983). Important Sites

The Thomas Creek Archaeological District (8SR338) containing Paleoindian sites is located in Santa Rosa County and is also listed on the National Register. Other important sites include the Page-Ladson site (8JE591) in northwest Florida; the Butler site (8SU2/8GI1), the Silver Springs site (8MR59) (Neill 1958; Hemmings 1975), Bolen Bluff site (8AL439) (Bullen 1958a), and Guest Mammoth site (8MR130) in north central Florida; the Harney Flats site (8HI507) in central Florida; and the Cutler Fossil site (8DA2001) in Dade County (Carr 1986).

Research Questions

In general, testing programs geared toward locating Paleoindian sites, which may be deeply buried, are needed. One strategy might be to retest known Archaic sites to determine if Paleoindian sites are underneath.

Chronology

Trends in artifact patterning (distribution and morphology) are largely undocumented within the period. Diachronic studies of artifacts from stratified sites will be required to address these questions.

Stratigraphically secure archaeological contexts containing diagnostic artifacts and datable materials are needed in order to reinforce or modify our beginning and ending dates for the Paleoindian period.

Technology

Settlement Patterns Questions