RETHINKING CULTURE HISTORY IN FLORIDA; AN ANALYSIS OF CERAMICS FROM THE HARRIS CREEK SITE (8VO24) ON TICK ISLAND IN VOLUSIA COUNTY, FLORIDA

Transcription

1 RETHINKING CULTURE HISTORY IN FLORIDA; AN ANALYSIS OF CERAMICS FROM THE HARRIS CREEK SITE (8VO24) ON TICK ISLAND IN VOLUSIA COUNTY, FLORIDA By CLIFFORD JOSEPH JENKS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA

2 Copyright 2006 By Clifford Joseph Jenks 2

3 TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT CHAPTER 1 INTRODUCTION ORANGE AND ST. JOHNS POTTERY, THE LATE ARCHAIC, AND TICK ISLAND The Ceramic Types of Harris Creek (8VO24) and Related Sites The Late Archaic and its Contextual Relationship to Tick Island and the Harris Creek Site (8VO24) The Archaeology of Harris Creek (8VO24) on Tick Island..21 The Geography and Environmental Setting of the Harris Creek Site (8VO24) ANALYTICAL METHODS Physical Attributes of Harris Creek (8VO24) Vessels.. 27 Stylistic Attributes of Harris Creek (8VO24) Vessels...31 AMS Dates from the Harris Creek Site (8VO24) on Tick Island ANALYTICAL RESULTS AMS Dating from the Harris Creek Site (8VO24) on TickIsland...37 Stylistic Results from the Harris Creek Site (8VO24) on Tick Island Techno-Functional Results from the Harris Creek Site (8VO24) on Tick Island...41 Summary of Results DISCUSSION AND CONCLUDING STATEMENTS Discussion of Results Social Implications of Ceramic Analytical Results from Tick Island.. 63 Conclusion

4 APPENDIX A B C D FIBER-TEMPERED AND DUAL-TEMPERED INCISED VESSEL MOTIFS SPONGE-TEMPERED INCISED VESSEL MOTIFS...69 VESSEL PROFILE FORMS FROM TICK ISLAND PHOTOGRAPHS OF INCISED VESSELS FROM TICK ISLAND 72 REFERENCES BIOGRAPHICAL SKETCH

5 LIST OF TABLES Table page 2-1. Data on AMS Assays of Soot Samples from Orange Fiber-Tempered Sherds from Middle St. Johns Valley Sites AMS Dates of St. Johns Ceramics from The Harris Creek Site (8VO24) on Tick Island Vessel Type and the Care Taken in Design Application Vessel Type and the Complexity of Elements Applied to Vessel Location of Rim Bands on Different Vessel Types Type and Function as Evident by Soot Deposits Form and Vessel Temper Form and Temper of Vessel as it Related to Function

6 LIST OF FIGURES Figure page 3-1. Examples of (a) bowl form; (b) flat wall form; (c) jar form Photographs of (a) boat shaped vessels form; (b) tray vessel forms Examples of designs with complex elements from Tick Island Examples of designs with simple elements from Tick Island Example of designs with banded prologues to decorative motifs Example of designs with banded epilogues to decorative motifs Profile Form and Pattern Design for Vessel #A Profile of Vessel A From Tick Island Profile of Vessel A From Tick Island Pattern design found on AMS dated vessel # Incised Surface Design on Vessel # Frequency of occurrence of preceding band counts on fiber-tempered pottery. Min: 3, Max: 10, Av: 4.8 lines, STD: +/ Frequency of occurrence of preceding band counts on sponge-tempered Pottery. Min: 1, Max: 2, Av: 1.8 lines, STD: +/ Frequency of occurrence of preceding band counts on dual-tempered pottery. Min: 1, Max: 8, Av: 2.35 lines, STD: +/ Wall thickness of Harris Creek (8VO24) vessel sherds. Max: 19 mm, Min: 2.5 mm, Avg: 9.37 mm, STD: +/ mm Wall thickness frequency for fiber-tempered vessel sherds at the Harris Creek Site (8VO24). Max: 14 mm, Min: 7 mm,avg: mm, STD: +/ mm Wall thickness frequency for dual-tempered vessel sherds at the Harris Creek Site (8VO24). Max: 19 mm, Min: 5.25 mm, Avg: mm, STD: +/- 2.65mm. 55 6

7 4-18. Wall thickness frequency for sponge-tempered vessels at the Harris Creek Site (8VO24). Max: 16 mm, Min: 2.5 mm, Avg: 8.22 mm, STD: +/- 2.4 mm Wall thickness frequency of cooking vessels at the Harris Creek Site (8VO24) Max: 18 mm, Min: 2.5 mm, Avg: 8.95 mm STD: +/ mm Wall thickness frequency of unsooted vessels at the Harris Creek Site (8VO24) Max: 19 mm, Min: 4.25 mm, Avg: 9.84 mm STD: +/ mm Wall thickness frequency of sooted fiber-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 12 mm, Min: 8 mm, Avg: mm STD: +/ mm Wall thickness frequency of unsooted fiber-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 14 mm, Min: 7 mm, Avg: mm STD: +/- 2.1 mm Wall thickness frequency of sooted dual-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 18 mm, Min: 5.25 mm, Avg: mm STD: +/- 2.7 mm Wall thickness frequency of unsooted dual-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 19 mm, Min: 6.5 mm, Avg: mm STD: +/ mm Wall thickness frequency of sooted sponge-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 16 mm, Min: 2.5 mm, Avg: 8.31 mm STD: +/ mm Wall thickness frequency of unsooted sponge-tempered vessel sherds at the Harris Creek Site (8VO24) Max: mm, Min: 4.25 mm, Avg: 8.08 mm STD: +/- 2.5 mm 59 A-1. Examples of fiber-tempered and dual-tempered incised vessel designs A-2. Additional examples of fiber-tempered and dual-tempered incised vessel design B-1. An example of an incised motif applied to a sponge-tempered vessel B-2. Additional examples of motifs found on sponge-tempered vessels...70 C-1. Examples of vessel profiles from the Harris Creek Site (8VO24) on Tick Island D-1. Photographs of dual-tempered and fiber-tempered incised ceramics from the Harris Creek Site (8VO24) on Tick Island..72 D-2. Photographs of sponge-tempered incised ceramics from the Harris Creek Site (8VO24) on Tick Island..73 7

8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts RETHINKING CULTURE HISTORY IN FLORIDA; AN ANALYSIS OF CERAMICS FROM THE HARRIS CREEK SITE (8VO24) ON TICK ISLAND IN VOLUSIA COUNTY, FLORIDA Chair: Kenneth E. Sassaman Cochair: Michael Heckenberger Major Department: Anthropology By Clifford Joseph Jenks August 2006 This study is an analysis of ceramic material found in northeast Florida at the Harris Creek archaeological site on Tick Island, which is located on the St. Johns River in Volusia County Florida. The primary purposes of this study were to better refine the current artifact-type chronology currently used, and to type artifacts by examining aesthetic style rather than intentional additives to pre-fired clay. Pottery from two chronological types were analyzed and compared for the purposes of this study; namely, the Orange and St. Johns ceramic variations. These pottery types are thought to be sequential in chronology, with Orange pottery being an earlier variant than St. Johns, but recent studies suggest that they may be more contemporaneous than originally thought. In order to test the possible contemporaneousness of the two vessel types, more radiocarbon dates were gathered from the St. Johns variant, and compared with dates from Orange ceramics which are readily available in the current literature. When a contemporary nature of the two ceramic types was confirmed, the study turned to account for stylistic variation between two ceramic types existing together at the same time. This analysis focused on 8

9 comparing the complexity and care of design application between the ceramic types, and found a correlation between design application and vessel functionality. Ceramics with more complex and careful design application were found to be less functional probably reserved for more ceremonial purposes and conversely, vessels with less complex and hastily applied designs were found to exhibit traits of a functional usage. 9

10 CHAPTER 1 INTRODUCTION Recently, the cultural-historic schemes of eastern and central Florida (Milanich 1994: ) have been called into question as a result of new AMS carbon dating of Orange fibertempered ceramics from the middle St. Johns valley (Sassaman 2003). The schemes, first established in the 1950s then further refined in 1972 by Ripley Bullen (Bullen 1972), have sustained a wide utilization and long adherence by archaeologists practicing in northeastern Florida. The results of new dating efforts may force Florida archaeologists to rethink ways in which sites and artifacts are analyzed, interpreted, and reported. The strict adherence to the stratigraphic contexts provided is problematic because they are highly unilineal in sequence. There have been positive steps to alleviate stratigraphic dependency through the use of radiometric dating and seriation, but the superimposition of archaeological assemblages into envisioned stratified deposits is still the primary method used to infer sequence. Often, these sequences are used to create regional chronologies where separate assemblages are arranged in an order of early to late culture-periods, with little consideration given to possible overlap. Through the rules of problematic normative culture-history, distinct assemblages are equated to units of distinct culture. Presently archaeology has been involved in the deconstruction of cultural-historic sequences formulated under the precepts of unilinearity and normative culture. This thesis is a contribution to refining the culture-history of northeast Florida. Formulated by Goggin (1952) and Bullen (1972), the cultural-historical sequence of the Late Archaic period (ca radiocarbon years before present [B.P.]) in the region is problematic. This era of early pottery development in the greater southeast is known in northeastern Florida for the traditions of Orange and St. Johns pottery as the key diagnostic features of ceramic development. Based on composite 10

11 stratigraphic data and limited radiometric assays, Bullen (1972) developed a cultural-historical sequence with four phases of the Orange Period, a transitional phase, followed by a St. Johns phase. However, recent radiometric dating, petrographic analysis, and intersite comparisons (Cordell 2004; Sassaman 2003) suggest that phases of the Orange period are contemporaneous rather than sequential, and a growing body of evidence suggests that the St. Johns tradition pottery is as old as ceramics of the Orange period. This paper has as its primary objectives to provide dates to further refine the culturalhistoric sequence of northeastern Florida; to better understand the more complex paste variation identified by Cordell (2004); and to provide alternative means for identifying cultural and temporal variation other than paste as it has been demonstrated that many of the petrographic features thought to be mutually exclusive to either Orange or St. Johns pottery types are not necessarily true. Many of the paste inclusions thought to belong only to St. Johns wares are now being found in the pastes of Orange pottery alongside the palmetto fiber (Cordell 2004). Style offers an excellent opportunity to understand culture and its change over time and space, for it is style that is a direct expression of culture. Using a stylistic analysis in conjunction with different temper types (rather than overall paste), newly acquired dates, form, thickness, and function (cooking/non-cooking), it is the hope here to offer a better picture of the culture-history at Tick Island s Harris Creek Site (8VO24) during the Late Archaic. This is a site-specific study, but used with the available literature and hopefully future studies of northeastern Florida s Late Archaic sites archaeologists will be able to make some well-founded generalizations concerning the regions culture history. 11

12 CHAPTER 2 ORANGE AND ST. JOHNS POTTERY, THE LATE ARCHAIC, AND TICK ISLAND As a preface to this analysis, it is important to provide some background perspective on the Orange and St. Johns ceramic types found at the Harris Creek site (8VO24). This will be followed by a historical look at previous work conducted at Tick Island, including its initial identification by archaeologists, and the archaeological activity on and off the site up to the present day. A generous amount of the information gathered concerning the sites archaeological history comes from research conducted by Lawrence Aten (1999). Also significant is the site s context in relation to geographic location, its relationship to the greater Southeast, and the time periods in which it was inhabited and utilized. The Ceramic Types of Harris Creek (8VO24) and Related Sites Orange ceramics: Orange pottery was first identified by Jeffries Wyman, who, in 1875, noted the use of palmetto fibers in the tempering of the vessels. It was through C.B. Moore s expeditions and detailed notes that clued archaeologists into the general context of fibertempered pottery in Florida. Moore s excavations in Florida demonstrated that the fibertempered pottery overlay a preceramic deposit and was, in turn, overlain by a more fine-grained check-stamped deposit of ceramic artifacts. James B. Griffin is credited with first noting the significance of the Orange period in Florida in 1945, defining what has been considered its most important type, which was the incised variety. He also noted it as similar to later St. Johns incised pottery, which have many of the same incised design elements and motifs (Bullen 1972). From the early 1950s to the early 1970s, Ripley Bullen began expanding the knowledge of this pottery type by dividing it into a chronological sequence based on form and style. The changes of these forms and style marked different sub-periods of Orange pottery and were used as 12

13 chronological markers. In his work, Bullen discusses Orange period sites as they relate to Orange chronology and not the time frame of fieldwork conducted at these sites. Bullen s chronology begins with what he dubbed Orange 1, ranging in age from approximately 4000 B.P. to about 3650 B.P. These wares were described as hand-molded pottery with thin walls ranging from 6-7 mm in thickness. They exhibited simple rounded lips and had the form of shallow, flat-based, and straight-sided bowls alongside some rectangular-shaped vessels. Some of these ceramics had lug like appendages, and all were untreated with surface decoration. Ceramics of this type were discovered at the Bluffton Site on the St. Johns River in the 1951 work of John W. Griffin in conjunction with Bullen and the Florida Park Service, where they conducted a large scale stratigraphic test. Here is where they found their first evidence of a plain fiber-tempered midden on top of 30 feet of pre-ceramic freshwater shell midden (Bullen 1972). Bullen interprets these vessels as being pseudomorphs of previously used wood trays, and baskets, a form which would be useful in the collection of Viviparous shells, a dominants species of St. Johns River middens. Orange 2 vessels were said to appear at 3650 B.P, and lasted until about 3450 B.P. The forms of these vessels were identical to that of the Orange 1 phase, however, these vessels exhibited incised surface treatment with motifs of concentric vertical diamonds with horizontal lines. Vessels from Tick Island are an exception to this phase s motifs, with renditions of spirals with background punctations, though this motif is rare, even on Tick Island itself (Milanich 1994). These vessels were also found at Bluffton and found with undecorated material, but not in any manner of stratigraphic superposition. Bullen mentions, though, that there is no noticeable temporal break, but a distinct difference in surface treatment and an increase in circular vessels as opposed to rectangular vessels (Bullen 1972). The surface treatment of the Orange 2 vessels at Bluffton includes incised concentric diamond motifs and a 13

14 Tick Island incised variant. St. Johns plain and check-stamped vessels were found in a disturbed overlaying context which also included some fiber-tempered pottery. At the Palmer site in Osprey was the second instance where Bullen found a plain fibertempered zone in the upper portions of a Late Archaic shell midden. The midden is described by Bullen as two parallel middens each 400 feet in length which are fused into an apex at the southern ends. Here, however, the majority of shells found at the midden belonged to marine species as opposed to freshwater shells. The maximum height of this midden site was approximately 5.3 m at maximum elevation. Many tests were placed in between the parallel middens and at its northern ends, and they were found to be sterile of material culture besides the shell refuse (Bullen 1972). Bullen attributes the sterility of these areas to the fact that the inhabitants would have been careful not to block the outlet. One major test and three minor ones were placed in the midden by Bullen and company, three being placed in the western part (Tests A-C), and one being placed in the eastern line of shell (Test D). Test A was, by far, the largest test, reaching a depth of approximately 4 m. The highest portion of this excavation block consisted of sand-tempered plain sherds, and those of the Norwood Plain variety. The next level of the block consisted of both Orange incised pottery and a sherd of St. Johns plain ceramic. In the next level between, approximately cm, the only ceramic content was Orange incised, and in the last arbitrary level at about 2 m below surface the only ceramic type to be found was Orange Plain. In the level where only incised Orange vessels were found, there was no curvature in the sherds, which indicated to Bullen that these vessels were flat-bottomed, straight-sided, rectangular containers. Incised motifs were difficult to identify on these vessels due to the fact that many of the sherds were very fragmented. However, it was ascertained to a certain degree that the incised designs on the vessels contained lines that were 14

15 straight and close together. One sherd did suggest a concentric diamond motif. The other tests conducted at Palmer yielded comparatively fewer sherds than were found in Test A. Test B, which was placed in the highest portion of the midden and dug to a depth of approximately 2 m, produced one Orange incised sherd at about 1 m below surface. Test C on the on the other side was dug to a little over 2 meters, and produced only one sand-tempered plain sherd in the higher levels of the test. In the eastern portion of the site, Test D was excavated to a little less than 2 m, and produced an Orange plain sherd as well as a sand tempered plain piece in the upper portion of the unit. At about 1 m below the surface there were four Orange plain sherds and two Orange incised sherds. Of importance concerning Palmer and its significance to the present study is that Bullen mentions that there does not appear to be any sort of break at the times of the introduction of, or changes in, ceramics at the site (Bullen 1972). To Bullen, the Orange 3 phase is typified best by the Summer Haven, Cotton, and South Indian Field sites (Bullen 1972). Summer Haven was considered important because it was one of the few Orange period sites located to the east of the inland waterway, making the environmental situation different from much of the St. Johns River where most Orange period sites are found (Bullen and Bullen 1961). Summer Haven is said to be largely destroyed by the mining of its shell for the construction of road beds. Bullen and associates worked on the eastern, undisturbed portion of the site which was being threatened by the widening of Route A1A. The midden in this portion was a little over 1 m in thickness. The excavations conducted at Summer Haven consisted of two adjoining 3x3 m excavation blocks (Bullen and Bullen 1961). This midden contained mostly Orange Plain and Orange Incised wares, though a Norwood Plain and a St. Johns Plain sherd were also found within their excavations (Bullen 1972). Aside from the ceramic assemblage recovered, the deposit was found to consist of shells with which were mixed 15

16 fluctuating degrees of charcoal, food bones, and occasional shell, bone, and stone tools (Bullen and Bullen 1961). The forms of Orange Plain and Incised vessels were a mix of straight-sided, round-mouthed, vessels with flat bottoms, and those that were not curved but had flat walls along with flat bottoms. Walls of these vessels ranged from 4-13 mm in thickness, and the bases were mostly 13 mm in thickness. The junction between the bases and walls were considered to be very thick, averaging 25 mm, and many of the walls were described to be wedge-shaped (Bullen 1972). Rims and lips of the Orange vessels from Summer Haven were described as simple and slightly rounded or wide and flat. Of the 354 rims examined by Bullen from Summer Haven, a little over half were decorated. The Summer Haven Orange incised vessels had a very wide variety of decorative motifs, according to Bullen. They were made up of different combinations of straight lines, ticks, and punctations. There were no curved lines, punctuated backgrounds, or concentric geometric figures found in examining these sherds. Designs described by Bullen include vessels with 6 to 9 parallel slanting lines, sometimes bordered by ticks, crosshatching, pendant hatched triangles, narrow hatched bands, and running frets (Bullen 1972). Rim decoration largely coincided with the overall surface treatment of the walls. These attributes differed enough from those listed by Bluffton and Palmer to warrant the designation of a new Bullen sequence. Summer Haven was also considered by the Bullens to offer a new view of the way of life of peoples of the Late Orange period in east Florida. The pottery is considered here to vary little when compared to the ceramics at Cotton and South Indian Field. Variations that do occur are considered to be minute decorative differences of little significance (Bullen and Bullen 1961), further justifying the designation of a third sub-period in his Orange phase sequence. The Cotton Site and South Indian field sites were said to be extremely similar in their ceramic composition 16

17 as well as other forms of artifact assemblages. Bullen identified the Orange 4 phase at the Sunday Bluff site, which he described as containing the first, pure Orange 4 deposits (Bullen 1972). Sunday Bluff is described as series of small shell middens located along a small tributary of the St. Johns River. In Bullen s test, the midden proper was a little under a meter in thickness and underlaid by soft sand containing pre-ceramic bifaces. The midden was overlaid as well by soft sand, and it contained Formative period artifacts such as St. Johns, and Deptford ceramics and more recent pottery and biface types. The midden itself contained Orange Plain and Orange Incised sherds exclusively, except for some St. Johns Plain sherds, usually in higher levels (Bullen 1972). Both Orange Plain and Orange Incised sherds from Sunday Bluff almost exclusively take the form of flat bottomed containers with thin walls, which are described to be approximately 6-9 mm in thickness. Traits of the Orange 3 phase are described as being absent from this site. Style at the Sunday Bluff site is described as being simplistic incised designs when the ceramics were decorated at all. Sunday Bluff, according to Bullen s descriptions, represented a terminal fibertempered phase of the Florida northeast culture region because of similarities between shapes and decorations of the, succeeding St. Johns Plain and Incised vessels. This situation plain chalky ware in otherwise fiber-tempered levels has been reported for both the South Indian Field and Cotton Sites (Bullen 1972). At a nearby site called the Colby Site Bullen states that Dr. Thomas Couchmore of Jacksonville had also found St. Johns sherds in fiber-tempered levels of his tests. Also found at the Colby site were, St. Johns Plain Sherds containing a little fibertempering (Bullen 1972). Bullen discusses a period which was dubbed Transitional, and believed that the Orange period ended by about 3,000 B.P. He further discusses a site which he believes to represent this 17

18 transitional period which he had identified. The Zabski site is described as originally being a one-period midden, and there was not a single fiber tempered sherd to be found. There were, however, St. Johns Plain, Incised, Pinched, Triangular Punctated, indented, and side lugged; Pasco and Perico Plain, Perico Linear Punctated; and sand-tempered plain sherds (Bullen 1972). The St. Johns vessels here are considered to be similar in form and decorative treatment to many of the Orange vessels, but the site lacked any fiber-tempered vessels, signifying to Bullen the close of the Orange period and the transition to a new phase of Florida ceramic assemblages. St. Johns phase ceramics: The St. Johns cultural sequence is described as, The Formative Stage, denoting a beginning of formal, settled communities, with the gradual development of more complex forms of political and religious community organization, is marked by a great deal more regional diversity than the earlier stages as (Milanich and Fairbanks 1980). It is believed that this diversity resulted from local adaptations to varied ecological conditions within the state (Janus Research 1996). The same seasonal pattern involving movement between the coast and river, said to be established during the Orange phase, is said to have continued during the early St. Johns period. This seasonal patterning, however, can be challenged, since there may have been more than one culture occupying differing environmental niches, one being on the coast and the other on the riverine interior. This may have earlier roots in a sedentary lifestyle that began in the Orange and continued through to the early St. Johns period. These can be marked in differences in ceramic assemblages and other artifacts between the interior and coastal settlements. Further work is warranted to determine whether or not coastal and interior sites represent a seasonal movement of a single people, or the habitation of differing environmental niches by two groups of people similar in some regards, but different in many others. 18

19 The St. Johns culture is said to be probably developed out of the fiber-tempered Orange culture found in the same region during the Late Archaic period, with great continuity in relation to the Timucuan speaking groups who lived in the region during the colonial period (Milanich 1994). The basic life-ways of St. Johns I culture are described as not too different from its Orange Late Archaic predecessor, who are said to have had a hunter-gatherer economy, while occupying villages and camps adjacent to many coastal and freshwater resources (Milanich 1994). Numerous archaeological surveys and excavations have significantly demonstrated that many Orange and St. Johns I period artifacts are often found at the same locales, and often at the same sites (Milanich 1994)! This is attributed to continuity of a cultural- chronological sequence, but may, in fact, represent a contemporaneous rather than a sequential relationship between the two ceramic types. The first recognized St. Johns Ceramic type is that of the St. Johns I ( B.P.) variety, and is described as a village ware with both plain surfaces and incised surface treatment (Milanich 1994). Forming techniques attributed to this ware describe a ceramic type that is coiled for all known wares. Some of the pottery is punctated or pinched, and the vessels, in a very few instances, exhibit the appearance of side-lug appendages. St. Johns Ia ( B.P.) describes village pottery that consists, for the most part, of undecorated wares found in association with Late Deptford and Swift Creek Pottery. St. Johns Ib ( B.P.) is described similarly as undecorated village pottery, often found with other ceramic types such as Weeden Island and Dunns Creek Red. These phases of St. Johns I pottery both are defined as a type by a surrounding context, and not by any traits inherent within the ceramics themselves. St. Johns IIa ( B.P.) is where the first checkstamped St. Johns wares were identified. This pottery is said to be found often in association 19

20 with Weeden Island variety ceramics. St. Johns IIb ( B.P.) is once again defined by its context, and not by any traits inherent within the ceramics themselves. This type is checkstamped pottery like its predecessor but found in association with Fort Walton and Safety Harbor pottery along with Southeastern ceremonial complex objects. Also noted of this ceramic type is a Mississippian influence at the sites in which they are found. St. Johns IIc is also described as check-stamped pottery, but in this case, in association with European artifacts in some middens and mounds. This pottery is said to belong to the various Timucuan-speaking groups described in European documentation (Milanich: 1994) The Late Archaic and its Contextual Relationship to Tick Island and the Harris Creek Site (8VO24) The Late Archaic ( B.P.): The most important development during the Late Archaic, as far as the subject of this paper is concerned, is the selective regional development of fiber-tempered ceramics throughout the Southeast. The word selective is used here because, although there was widespread population growth and a profusion of sites in the region, ceramics only seemed to be innovated and utilized in certain core areas. These core areas were along the St. Johns River with Orange Phase ceramics, in South Carolina with the Stallings Island variety of fiber-tempered pottery, and Thoms Creek a little north of the Stallings Island area (Sassaman 1993). With the widespread occupation and interregional interaction of the time, people would most certainly be aware of this innovation in cooking and serving technology, but for some reason its spread was being hindered. This once again suggests the possibility that some vested interests were at work to stifle this innovation. It also lends additional support to a main theme of this paper; that is, that not all traits and innovations will automatically be adopted within a culture, regardless of their functionality or expedience. 20

21 The Archaeology of Harris Creek (8VO24) on Tick Island Tick Island, bordered by Lake Dexter, Lake Woodruff, and Mud Lake on the 440 km stretch of the St. Johns River (Jahn and Bullen 1978), was initially identified during the 19 th century expeditions of Clarence B. Moore on the middle portion of the river. The Harris Creek Site (8VO24) is a ceremonial shell midden with materials ranging from the Middle Archaic period ( BP), up to the St. Johns IIb ( BP) period of Florida cultural chronology (Jahn and Bullen 1978). It was described by its initial investigator as having a wild appearance, covered as it is with gnarled live oak and towering palmetto with trailing vine and tangled undergrowth. (Moore 1892a in Aten 1999). Moore reported his fieldwork conducted in February, March, and April of 1891 at the island and its northernmost site (8VO25), but it wasn t until 1893 that he discussed the main southern site, the Harris Creek Site (8VO24), which is the subject of this current study. Moore s initial description of the Harris Creek site (8VO24) referred to a circular heap of shell converging to an apex at the center. (Jahn and Bullen 1978). The excavation conducted at the Harris Creek Site (8VO24) was a 2.1x1.2 m test block, which was excavated to a depth of 2.7 m. In his excavation block, Moore encountered a layer of humus in the first.4 m, and in the remaining 2.3 m, was a thick layer of shell deposit with a heavy concentration of plain and decorated fiber-tempered pottery (Moore 1892 in Aten 1999). A review of further documentation by Moore revealed that there were additional excavations that he conducted at the Harris Creek Site (8VO24). In these excavations, Moore uncovered bone tools, and projectile points alongside the ornamented pottery, at a depth of m. Another pit excavated by Moore revealed four burials, one of which is described as flexed and directly over a fireplace, but gave no evidence of contact with flames (Moore 1892 in Aten 1999). 21

22 Francis Bushnell did additional work at Tick Island in 1959, and furthered the archaeological knowledge base concerning it by mapping the island in relation to its cultural and natural features (Aten 1999). During his work there, he conducted controlled surface collections, and divided the site into four separate areas differentiated by the letters A-D, which are still referred to today when discussing the site. Of particular significance was Area D, which consisted of burials beneath shell at a depth of approximately 1.5 m below the surface. One of these skeletons was found to have several projectile points embedded in it. Several of the skulls have been interpreted to have been crushed in a violent manner. It is possible, however, that they were crushed by the weight and pressure of the overlying sediment over a long period of time. Area A was where the greatest diversity of ceramic types were found, ranging from Orange period ceramics to the St. Johns II period. Area C of Bushnell s observations was a mortuary segment of the site with three small inland shell mounds, each containing a single burial. The only area that was not as substantially destroyed as the other areas was Area D, and Bushnell conducted controlled excavations of this area. He described it as follows: A striking hill or rise which might at first be mistaken for a temple mound structure. In this rise, sherds of the St. Johns Period are almost absent, being found only on the direct surface. Orange plain appears only at a shallow depth, with a possible preceramic region showing up at about two feet. Later work at Tick Island includes controlled surface collections and observations made by Otto Jahn between 1964 and 1969 at the Harris Creek Site (8VO24). In the 1980s, there were also core samples taken from archaeological deposits and well-preserved organic materials (Aten 1999). The site s integrity has been highly compromised as of today, since the island was heavily damaged by shell mining conducted throughout the 19 th century up to the 1970s. Tick Island is 22

23 heavily inundated and just a little shell is exposed over the water s surface. Luckily, salvage excavations have preserved a great many of the artifacts from the site, and they are available for study at the Florida Museum of Natural History in Gainesville, Florida. New AMS dating techniques make it possible to determine accurate dates from just a small amount of soot deposits from ceramics used over an open flame. This valuable technology now makes it easier to get an idea of context concerning these ceramics since the sherds can be more precisely dated. This advance has allowed archaeologists to make new interpretations concerning sites like Tick Island in which a stratigraphic context is no longer available. Recent work done concerning Tick Island, was the dating of northeastern Florida s Orange ceramics to obtain a more refined and accurate chronology regarding these types. The majority of these sherds came from Tick Island and Mouth of Silver Glen Run (Sassaman 2003). These new dates have inspired further research, included here, into the nature of Orange-period ceramics as they relate to surrounding St. Johns ceramics, as well as currently held cultural-historic beliefs concerning Archaic and formative Florida ceramics. The Geography and Environmental Setting of the Harris Creek Site (8VO24) The Harris Creek site (8VO24) is located on Tick Island in the St. Johns valley between Lake Dexter and Lake Woodruff. The valley at this location is broad compared to the areas to the north and south (Jahn and Bullen 1978). Tick Island was described at the time of Jahn and Bullen s report as being forested, except for marsh on the western end, with a wooded swamp along the edges and a pine and palmetto forest in the central higher elevations. A wooded swamp to the north of the island runs across to the mainland. This geographic description was made in 1978 by Otto Jahn and Ripley Bullen, but land alterations over the years since then could have changed the site s present physiographic and geographic appearance. 23

24 More current descriptions of the site s geography include references to a 16-Km expanse of streams, shallow lakes, vast marshes, and tree hammocks (Aten 1999). The Harris Creek Site (8VO24) itself is located on the southeast portion of the island. It is described as having very little current flow, and a good habitat for the Viviparus georgianus freshwater snail, the dominant species of shell found at Tick Island. The predominant soil type found across the floodplain at Tick Island is a Terra Ceia Muck, which is a highly organic and hydrated deposit. Elevations in the floodplain surrounding Tick Island and its associated sites are estimated to be below 1.5 m (Aten 1999). This floodplain soil type plays host to a wide range of vegetation including; saw grass marshes, cordgrass marshes, and hammocks containing a variety of hardwood tree species. The center of the island has the area s highest elevation of 3.3 m and is defined in county soil surveys as nearly level with poorly drained, Farmton fine sand. This type of sand is described as being developed on top of clayey and silty marine sands that underlie the floodplain deposits around the island s borders. Between the aforementioned soils found in Tick Island s environmental setting is a third type that transitions between the two as the island slopes into the flood plain. This soil type is defined by county soil surveys as Tuscawilla fine sand which, logically, is a mixture of the soil types on the island and in the floodplain. Lowland hardwood hammock vegetation is considered more typical on this transitory soil type (Aten 1999). The surrounding middle and upper reaches of the St. Johns River valley are well-known for a series of shell midden sites. Tick Island is probably the largest and best-known. A postglacial rise in sea level during the Middle Holocene backed up the St. Johns River to leave a series of shallow lakes, oxbows, and dead rivers. This was an ideal environment for freshwater shellfish such as the Viviparus georgianus, mentioned previously, as well as large apple snails (Pomacea paludosa), and an array of mussels and freshwater clams (Jahn and Bullen 1978). This 24

25 change in environment opened up a cornucopia for aboriginal peoples of the region to exploit, and is most likely responsible for the utilization of Tick Island as well as the various other midden sites found in the St. Johns River valley (Jahn and Bullen 1978). Lab Number Beta- Table 2-1 Data on AMS Assays of Soot Samples from Orange Fiber-Tempered Sherds from Middle St. Johns Valley Sites Site Sample* Material 8LA1-12 soot Beta LA1-6 soot Beta Beta Beta Beta Beta LA1-27 soot 8LA28-2 soot 8LA24-21 soot 8LA24-1 soot 8LA soot Calibration Measured C13/C12 Conventional Intercept Intercept 2- Sigma C14 Ratio C14 (Cal BC) (Cal BP) (Cal Age (BP) (0/00) Age (BP) BC) 3690+/ / / / / / / / / / / / / / Sigma (Cal BP) *8LA-1 Mouth of Silver Glen Run; 8LA28-Mosquito Hammock; 8VO24-Tick Island Sassaman 2003: New AMS Dates on Orange Fiber-Tempered Pottery from the Middle St. Johns Valley and Their Implications for Culture History in Northeast Florida. The Florida Anthropologist 56(1):

26 CHAPTER 3 ANALYTICAL METHODS All of the ceramics studied for this analysis were cleaned, processed, and curated at the Florida Museum of Natural History long before this analysis was undertaken. These ceramics were collected over many years during several expeditions made to Tick Island and the Harris Creeks site (8VO24), and have long since resided at the museum. Since this analysis was focused on a vessel unit of study, the sample chosen for examination was confined to rim sherds. Vessels chosen to be studied included Orange and St. Johns Incised vessel sherds, typed as such based on macroscopic and tactile properties of the paste. Although one of the primary concerns of this study was style, the St. Johns plain vessels were chosen in order to have something to compare vessels with both decorative and non-decorative surface treatments as they relate to issues of form and function. There were no Orange vessels that were undecorated in order to make a similar selection. Analyzed vessels were coded according to several physical and stylistic attributes to be used to compare and contrast the different ceramics. Physical attributes included temper, body thickness, form and function. Stylistic attributes were coded for presence or absence of decoration, and stylistic elements -- both simple and complex. Simple elements are considered those that take one stroke of a stylus such as a horizontal line, vertical line, diagonal line, or tick mark. Complex elements were those combining simple elements to provide different shapes and figures. Certain vessel sherds of the St. Johns type that had soot deposits present on the exterior were chosen in order to get direct dates associated with that pottery to help further refine the cultural-chronological sequence. Orange vessels were not included in the AMS dating, since dates have already been acquired from several Orange sherds within the region (Sassaman 2003). The reason for studying physical attributes in conjunction with stylistic ones is that the 26

27 present study believes style and function are not mutually exclusive categories. Stylistic traits applied to vessel surfaces may in fact be related to the type of function for which the ceramic may have been intended. Differing vessel forms, although more conducive to different functions (Hally 1986), can be considered a stylistic and functional trait. Style has long been used as a descriptor of exterior decorative surface treatment, but may not be so easily separated from function. It is hoped that the following evidence provided by this analysis will demonstrate that. Physical Attributes of Harris Creek (8VO24) Vessels Temper: This attribute refers to the inclusion of extra materials in the clay before firing a vessel in order to alter the performance characteristics or other physical properties of the vessel. As a noun, it refers to the components within the paste of a vessel assumed to have been added intentionally, and as a verb, it refers to the action of adding components not naturally occurring in the clay (Rice 1987: 406). These tempering agents were used to change the properties of the clay during the various stages of its production throughout and after its firing. Tempers used over various times and locations, have varied greatly according to the intentions of the potter. Agents used to temper vessels have included plant: grass, or plant fibers, chaff or straw, cattail fuzz, and plant silica. Tempering materials derived from animals have included shell, sponge spicules, and dung. Mineral tempering has been used as well, and has included various types of crushed rock, such as limestone, sandstone, andesite, trachyte, basalt, sand (as it is composed of quartz), and volcanic ash. Material of human origin, such as former potsherds and brick remnants, have also been cited as examples of tempering elements used by some potters (Rice 1987:407). Of key significance is the clear evidence that tempering agents were deliberately chosen by potters for specific purposes. This leads to the obvious question of what those purposes might have been (Rye 1976). What technological advantages did one tempering material provide over another? 27

28 In order to determine the tempering agents used in the ceramics examined for this study, all specimens were viewed microscopically. Before viewing, a sample was taken from the least intrusive portion of the sherd in order to avoid damaging the surface treatment of such specimens. It was important to collect as much information as possible while incurring the least possible damage. It is often necessary to make a break in the ceramic before viewing it microscopically because when they are taken out of their original archaeological context, many of the surrounding minerals and debris end up mixing with the exposed core. In order to minimize the risk of including material from the surrounding environment in which the sherd was found, a break was needed to view the core as it was at the time of its original deposition into the archaeological record. Tempering agents seen in the sherds studied at Harris Creek (8VO24) include sponge spicules, and plant fibers (presumably from Spanish moss and palmetto plants). Both agents have been noted as creating different qualities within a vessel intended for a particular type of use. Experiments conducted on organic tempered ceramics have revealed some of the advantageous qualities of using plant fibers as an additive to clay (Skibo, Schiffer, and Reid 1989). The main advantage that fiber-tempered vessels appear to have provided was an ease of portability because of their light weight, which would have made them an ideal pottery type for mobile or semimobile hunter-gatherer groups. Fiber-tempered pottery, as opposed to those vessels tempered by sand, has been shown to be up to 34% lighter in overall weight. This reduction in weight is also credited with making the vessels less likely to break when accidentally dropped. Another advantage of porous vessels is that the pores particularly larger pores work to inhibit the propagation of cracks as a reaction to temperature change. If a crack begins, it is halted by the pore (Rye 1976). Disadvantages of fiber-tempered pottery are that they are more susceptible to 28

29 abrasion than mineral or un-tempered vessels due to the porous nature of their pastes. Organictempered vessels also did not have efficient heating effectiveness. Because they were thick and porous, it was difficult for heat to transfer from the exterior to the interior. It is of interest to note, however, that these vessels were still used over an open flame to a lesser extent in spite of this inefficiency (Sassaman 2003). Although experimental studies have not been done on sponge-tempered vessels to the extent that they have on those of organic-tempering, one can infer that these would have similar properties to either the untempered or mineral-tempered vessels. (Skibo et al. 1989). These vessels would have been more effective for heating, less susceptible to abrasion, and heavier. Wall thickness: The attribute of wall thickness is related to a containers size, and intended use (Rice 1987:227), as well as the strength of the clay being used to craft a particular vessel. Wall thickness for this analysis was measured three centimeters below the lip with a pair of calipers. The thickness of a wall is said to affect three main aspects of mechanical performance; namely, thermal conductivity, flexural strength (breakage load), and resistance to thermal shock (Braun 1983). It has generally been thought that thicker vessels are more appropriate for storage purposes since a thicker base is said to increase stability as well as keep moisture in or out of the vessel (Rice 1987:227). Thicker walls are also said to be stronger and more resistant to sharp blows during pounding, stirring, or mixing. They serve as a disadvantage during cooking due to the fact that it takes much longer for them to conduct heat, as opposed to their thinner counterparts. Thin walls are quicker conductors of heat, and they benefit their users by increasing the vessel s potential for resistance to thermal shock (Braun 1983). The disadvantage of these thin walls is their decreased flexural strength, but this weakness can be counteracted by the curvature of the vessel (Braun 1983). It is thought that the smaller the radius 29

30 of curvature of any given wall, the higher its resistance will be to mechanically induced fracture. Thick walls are said to be useful in transfer functions because the walls are slow to conduct heat from the inside out, making the vessel easier to grasp, but their disadvantage is their heavier weight which makes the vessel difficult to handle despite being cool to the touch (Rice 1987: 228). This disadvantage could, however, be counteracted by thick-walled, fiber-tempered vessels in which the porosity would make them about a third lighter than most vessels tempered with other agents (Skibo et al. 1989). Wall thickness from the Harris Creek (8VO24) vessels, was measured with calipers. All thickness measurements of the ceramics studied here were taken at 3 cm below the lip of each sherd. Rim form: Form was recorded for each sherd using a contour gauge, a tool made up of wire bristles in which an object s form is duplicated when pressed into the bristles, leaving a replica of the object s shape within the gauge. After the contour gauge was utilized, the shape was traced onto graph paper from the shape formed in the bristles. The resulting profiles of each vessel were examined and then coded as a particular form based on the attributes of each profile. Different forms are seen as being conducive to particular functions. Certain physical and morphological features determine the limits of a vessel s mechanical performance characteristics, and can therefore be used with a good degree of confidence in making certain inferences about their use (Hally 1986). Five vessel forms were identified and recorded from the profiles studied here. These included (Figure 3-1) bowls, trays, boat-shaped vessels, and flat-walled vessels. It was difficult to draw an accurate vessel profile for trays and boat-shaped vessels due to the inconsistencies on the surface of their rims. These two categories were, therefore, identified directly from the sherds themselves (Figure 3-2), and not from the drawn profiles. These categories were subjectively categorized based on sherds that exhibited 30

31 characteristics in a consistent manner, but were not readily identifiable from previously identified forms (Hally 1986). Boat-shaped vessels were so named, because their walls resembled the sidewall of a canoe, curving upward and inward to an end point, or presumed convergence point, with the other wall. A trait not recorded for these boat-shaped vessels, but observed during this analysis, was that on the wall near the convergence points there was almost always a purposely drilled hole, which is interpreted here as a suspension hole to place the vessels over an open flame. Some vessel sherds were placed in the category of trays because of the sharp angle they made when placed against a flat surface to interpret profile form. Any other sherds, except the boat-shaped variety, would measure closer to a 90-degree angle where the rim intersected with a flat surface. Flat-walled vessel sherds were so classified because they lacked any curvature down the wall when their rim was placed on a flat surface. These types of walls may have met with other walls of the same vessel to form rectangular containers. Such containers have been recorded among undecorated vessels of the first designated Orange phase, but appear as decorated vessels, since all of the Orange vessels studied at Tick Island had incised decorative motifs. Undecorated Orange vessels were absent from this mortuary site, but there was a large group of St. Johns vessels that were undecorated and selected as units of study. Stylistic Attributes of Harris Creek (8VO24) Vessels After the physical attributes were identified and recorded, the focus of the analysis turned to the stylistic tendencies of the ceramics. Style, in this context, has been defined as variation in image design, as distinct from designs that are considered to be representational, nature-based, and realistic, which constitute renditions of subjects seen by the potter (Rice 1987: 247). Other vessels can take on designs that are considered abstract, iconic, or geometric, in which the style has been reduced to a selection of particular features considered in some way as 31

32 essential or basic (Rice 1987: ). Stylistic designs on incised vessels at the Harris Creek site (8VO24) were all geometric in nature and applied by the potters to achieve various stylistic motifs. All aspects of pottery creation can be construed as style, since the potters make a set of decisions related to such characteristics as form, temper, wall thickness, and surface treatment, all of which are expressions of the activities and symbolism important to the culture at large. The term culture will be used here to refer to the active use of shared symbols among a given people. These symbols can be manipulated and changed over time in the course of rebellions and assimilation in the ever-present process of culture negotiation that takes place among sentient beings that utilize symbols When doing the stylistic analysis, sherds were first coded for the presence or absence of decoration. Vessels sherds were then analyzed for the type of design that was applied upon their surfaces. Pottery design at Tick Island have been categorized as either incised or check-stamped, however, check-stamped sherds have been excluded from this study because there were too few specimens available (only two) to draw any significant conclusions about stylistic or technological tendencies among these vessel types. In analyzing the quality of design application on the vessels, a subjective category was coded for. It was noted in the initial examination of these sherds, that several vessels had their design applied with great care and detail (Figure 3-3), while others were seen to have designs applied in a haphazard fashion with little concern for quality (Figure 3-4). Obviously, these categorizations are based on subjective value judgments. However, when the two categories are compared, the distinct differences become apparent and would certainly seem to support these inferences. Upon this initial observation it was apparent that new questions would have to be raised to account for these differences in the degree of care taken in the application. 32

33 This study coded for simple and complex design elements. A simple element is defined as involving a one-directional stroke of a stylus creating a horizontal, vertical, or diagonal line, or punctation. A complex element is defined as a series of these simple elements to form a patterned design or a shape. It was also found during the analysis that many of the designs had either a prologue of horizontal bands that preceded the main design motif (Figure 3-5), or an epilogue that was drawn underneath the design motif (Figure 3-6). Incised sherds were coded for lines preceding the design and lines concluding the design, and a count was taken of the number of these bands and recorded for each sherd. It was of interest here whether these preceding or concluding lines had any impact on the overall motif of the vessel in terms of any possible grammatical rules concerning design application below or above the band or bands. As it progressed, this analysis focused more specifically on a study of incised treatments found at the Harris Creek Site (8VO24), since this was the predominant type of stylistic expression occurring on all of the vessels regardless of paste differences. AMS Dates From the Harris Creek Site (8VO24) on Tick Island Many sponge-tempered incised sherds, as well as some in the fiber-tempered and dualtempered categories, exhibited soot deposits on the exterior surface indicating use over an open flame. Sherds were first coded for the presence or absence of this soot, which had two advantages. The first being the strong evidence it provided for cooking function, since it can be reasonably inferred that vessels placed over an open flame were used to heat food. The second advantage is that the presence of soot thanks to recent advances in radiometry makes it possible to ascribe direct dates to ceramics even when found out of their original archaeological context and cultural deposit. Direct radiometric dates can now be taken from sherds with only a 33

34 minimal amount of soot present. This is because improvements in accelerated mass spectrometry (AMS) technology have made it possible to achieve very accurate dating from even a pin-point sized sample of carbon (Sassaman 2003) This type of dating has proven to be very successful in its utilization throughout the Southeast and other areas. In the course of this study, samplings of soot deposit taken from five St. Johns cooking vessels from the Harris Creek Site (8VO24) on Tick Island were sent to AMS labs for dating. These dates were to be used to refine established cultural historic chronology at Tick Island specifically, with possible implications for the broader region. These dates, however, should not be over-extrapolated to represent a whole Florida sequence. In fact, one of the objectives in acquiring these dates was to support this paper s contention that the Florida cultural-chronological sequence may be a case of over-extrapolation of dates from highly separated sites and regions, and that pottery forms and types may have changed in different ways over time at these sites. 34

35 (a) (b) (c) Figure 3-1. Examples of (a) bowl form; (b) flat wall form(c) jar form. (a) (b) Figure 3-2. Photographs of (a) boat-shaped vessel form; (b) tray vessel forms. Figure 3-3. Examples of designs with complex elements from Tick Island. 35

36 Figure 3-4. Examples of designs with simple elements from Tick Island. Figure 3-5. Example of designs with banded prologues to decorative motifs. Figure 3-6. Example of designs with banded epilogues to decorative motifs. 36

37 CHAPTER 4 ANALYTICAL RESULTS Results from this analysis will be discussed beginning with the new AMS dates obtained from the selected sherds at Tick Island in order to provide a chronological context in which to better position the sherds in culture history. Following this discussion of the new AMS dates, will be a description of the results of the analysis of the stylistic characteristics of the specimens. Concluding the discussion will be an analysis of the techno-functional aspects of the ceramics. AMS Dating from the Harris Creek Site (8VO24) on Tick Island The AMS dating from vessel sherds tempered only with sponge spicules at Tick Island yielded dates as recent as /- 40 B.P., and also yielded a much earlier date of /- 40 B.P. (Table 4-2). As far as the cultural-chronological schemes in current use are concerned, this places these sponge-tempered vessels in alignment with Bullen s Orange I sub-period of the Late Archaic at Tick Island. The most recent dates that came from spicule-tempered vessels at the Tick Island site placed them in the Orange 5 phase of the Late Archaic. This new data makes it necessary to look at Orange and St. Johns vessels in a new light. Vessel # A (2700 +/- 40 B.P.): This sherd was a sponge-tempered incised vessel containing only sponge spicules in the paste. The wall thickness of this vessel sherd was approximately 13 mm. The vessel was incised in a hasty fashion, and contained only simple elements. The simple elements incised into the vessel sherd s surface were horizontal and vertical lines, and the overall patterning was concluded with one horizontal band. (Figure 4-7). Vessel # A (2700 +/- 40 B.P): The assay of soot from this vessel sherd matched the same AMS date of the previous one, and was a bowl form. It was a sponge-tempered incised vessel, and the wall thickness of this vessel was approximately 8 mm. The design incised on the 37

38 exterior of the sherd was interpreted to be hastily applied, but contained both simple and complex elements. Simple design elements included horizontal lines, diagonal lines moving from the bottom right side to the top left side, and diagonal lines moving from the bottom left side to the top right side. These simple elements came together to form one complex element on the vessel, which was a triangle. The vessel design was preceded by one horizontal band concluding the decorative motif (Figure 4-8). Vessel # A (2940 +/- 40 B.P.): This vessel sherd was an undecorated spongetempered ware that took on the form of a bowl. It was purely tempered with sponge spicules since microscopic analysis did not indicate the present of burnt out plant fibers. The vessel had a wall thickness of about mm. (Figure 4-9). Vessel # (3370 +/- 40 B.P.): This was a boat-shaped vessel sherd with a wall thickness of approximately 9 mm. This particular vessel, like the previous two, was only tempered with sponge-spicules. This vessel however, unlike the more recent ones, had a careful design application, although the design included only simple elements. These were horizontal lines as well as diagonal ones that went from bottom left to top right. This vessel, unlike the previous ones, had lines preceding the overall design rather than concluding it. There were two horizontal bands running parallel to the rim (Figure 4-10). Vessel # (4110 +/- 40 B.P.): This oldest of the vessels studied was boatshaped in form, and was tempered only with sponge spicules. It had a wall thickness of approximately 8.25 mm, but the condition and form of this specimen also precluded a measurement of orifice diameter or the creation of a profile drawing. The design on this spongetempered vessel was carefully applied and included both simple and complex elements contributing to the overall motif. Simple elements included diagonal lines running in both 38

39 directions, as well as horizontal lines. Complex elements for this vessel included what have been defined here as closed chevrons, as well as triangles, and a reverse Y shape. This vessel had two bands running parallel to the rim and preceding the overall design motif (Figure 4-11). Stylistic Results from the Harris Creek Site (8VO24) on Tick Island Vessel type and design application: There was definitely a significant difference as to the manner in which designs were applied to vessels according to type. (Table 4-3). Of the 19 purely fiber-tempered vessel sherds observed, none had a hasty application of the design while all 19 had what was considered to be a carefully applied motif. When looking at the 50 vessel sherds tempered with, both plant fibers and sponge spicules, three (6%) had a hasty design application, and 47 (94%) had designs implemented with care and precision. Among the 76 vessel sherds purely tempered with sponge spicules, 71 (93.42 %) were considered to have a hasty application, while only 5 (6.58%) were considered to have designs that were applied with care. Design element complexity and type: Fiber-tempered and dual-tempered vessel sherds had a much lower number of stand-alone simple elements than those with simple elements coming together to form more complex designs. (Table 4-4). Out of the 19 fiber-tempered vessel sherds examined 6 (31.58%) had simple elements alone, while 13 (68.42%) had simple elements which combined to make a more complex design motif. The dual tempered vessels were similar in their percentage with 15 (30%) of the total 50 vessel sherds having only simple elements, and 35 (70%) having complex elements. Out of the 76 sponge-tempered vessel sherds, 37 (48.68%) had stand-alone simple incised elements, while 39 (51.32%) had simple elements worked together to form more complex elements for the overall vessel motif. There was absolutely no presence of bands on the epilogue of either dual-tempered or 39

40 fiber-tempered vessel sherds, and epilogue bands never exceeded one. This was a trait reserved totally for sponge-tempered vessel sherds, since all 47 that exhibited a design epilogue had a pure sponge-spicule tempering. (Table 4-5). Sponge-tempered vessel sherds did have minor representation in those vessels with a banded prologue, with five (14.71%) of the total 34 vessels with prologues being St. Johns. The fiber-tempered and dual-tempered vessel sherds made up the rest of these motif prologues with a representation of six fiber-tempered vessel sherds (17.65%), and 23 dual-tempered vessel sherds (67.64%). Out of the total 52 incised sponge spicule vessel sherds with banding, 47 (90.38%) had banding below the motif and 5 (9.52%) had it above the motif. Out of the total sample of 81 incised vessels with some sort of banding, 47 (58.02%) had banding below the design and 34 (41.98%) had banding above the design. It is apparent from reviewing the data that bands preceding a design were a trait designated mainly for fiber-tempered (Figure 4-12) and dual-tempered (Figure 4-14) vessel sherds. There was a far greater range in the number of bands applied as well as the frequency of these bands occurring on these types of ceramics. They were found to a far lesser degree on some of the sponge-tempered vessel sherds, and their range of band numbers greatly decreased to one to two bands (Figure 4-13), as opposed to the fiber-tempered prologue bands whose numbers ranged from one to ten bands and the dual-tempered vessel sherds whose bands ranged from one to eight. As far as concluding design bands are concerned, this was a trait reserved specifically for sponge-tempered vessel sherds, and there was no variation in band numbers. All of these sherds were concluded with just one band running horizontally below the main design motif. There were a total of 47 sponge-tempered vessels with a horizontal band running below the rim's design. 40

41 Techno-Functional Results From the Harris Creek Site (8VO24) At Tick Island Temper: Of the total 145 vessel sherds (Table 4-6) from the Harris Creek Site that were examined, there were 19 (13.1%) specimens that were tempered only with fiber, and 76 (52.42%) that were tempered only with sponge spicules. Dual tempering was represented with 50 (34.48%) vessel sherds. There was a very substantial use of two tempers in vessel sherds that would normally be classified as Orange ceramics according to currently utilized macroscopic properties. Function: On looking at function in relation to temper (Table 4-6), it was observed that a large proportion of specimens of fiber-tempered and dual-tempered vessels showed no indication of soot deposits. Out of 19 fiber-tempered vessels examined only five (26.32%) showed evidence of use over an open-flame, while 14 (65.22%) showed no such evidence. On the other hand, a much greater proportion of the sponge-tempered specimens contained soot as evidence of use over an open flame. When looking at the 50 dual-tempered vessels 19 (38%) showed clear evidence of soot deposits while the other 31 (62%) showed no signs of such deposits. Of the 118 sponge-tempered vessel sherds examined, 75 (63.56%) showed evidence of use over an open flame, while 43 (48.87%) did not. Form: The majority of fiber-tempered and dual tempered sherds took on the form of either bowls or flat-walled vessels (Table 4-7).There was a very small minority of these vessels that had taken the form of a jar. Jars were not included in the analysis of form since their presence was so small that they were unable to relay any significant information. Of the total 19 fiber-tempered vessel sherds under analysis or the 48 dual-tempered vessels, there was no presence of the boat-shaped form. This was found only among the sponge-tempered vessels. There were 6 (31.58%) bowl shapes, and 13 (68.42%) flat-walled vessel sherds making this latter 41

42 form the majority of fiber-tempered vessels. The 48 dual-tempered vessels consisted of only bowls and flat walls. The boat-shaped vessels, as stated earlier, were only found among spongetempered vessels. Of the 48 dual-tempered vessels, 14 (29.2%) were found to take on a bowl form, and 34 (70.8%) were flat-walled vessel sherds. The great majority of dual- tempered vessels were those consisting of flat walls as shown by the statistics. These percentages are very similar to the forms making up purely fiber-tempered vessels. When viewing the 111 spongetempered vessel sherds, 15 (13.51%) of them were the newly identified boat-shaped vessels, 57 (51.35%) were bowls, and 39 (35.14%) were flat-walled vessels. The emphasis shifts from flatwalled vessels in the fiber-tempered type, to bowl forms in the sponge-tempered type. However, flat-walled vessels still make up a significant number of the sponge-tempered type vessel sherds. Boat-shaped vessels were unique to the sponge-spicule paste. Form and function: There were a total of five fiber-tempered vessel sherds that had soot deposits present on the exterior (Table 4-8). There were three (60%) sooted bowls and 2 (40%) sooted flat-walled vessels. There were a total of 14 unsooted, fiber-tempered vessel sherds under study, and of these, 3 (21.43%) were bowl-shaped, and 11 (78.57%) were flat-walled. There was a significantly larger number of unsooted flat-walled vessels as far as this sampling is concerned. When looking at dual-tempered vessel sherds, there was once again a lack of boatshaped vessels as stated previously. Of the 18 sooted, dual tempered vessels there were seven (38.89%) bowl shaped vessel sherds, and 11 (61.11%) flat-walled pieces. When looking at nonsooted, dual-tempered vessel sherds there were seven (23.33%) bowl profile forms, and 23 (76.67%) flat-walled forms out of a total of 30 vessel sherds in this category. As far as form and function are concerned, sponge-tempered boat-shaped vessels were fairly evenly divided between those that were used over an open flame, and those that were not. Of the total 15 boat- 42

43 shaped vessel sherds that were analyzed, eight were found to have soot deposits present on the exterior, while seven of the sherds lacked this feature. Of all 71 sponge-tempered, sooted vessel sherds, eight (11.27%), as stated previously, were boat-shaped, 36 (50.7%) were bowls, 27 (38.03%) were flat-walled. Sponge-tempered vessel sherds that lacked soot deposits totaled 40, and seven (17.5%) of these were considered to be boat-shaped vessels, 21 (52.5%) considered to be bowl shapes, and 12 (30%) considered to be flat-walled. Wall thickness: Overall, the average wall thickness of vessels from the Harris Creek site (8VO24) on Tick Island was / mm. (Figure 4-15). When isolating fiber-tempered vessels by themselves (Figure 4-16), this average for wall thickness increases to /- 1.97mm, and when looking at sponge-tempered vessels, the average wall thickness decreases to 8.22 mm, not substantially less than the overall average of vessels from the Harris Creek site (8VO24) but a good deal less than the fiber-tempered vessel sherds. When looking at vessel sherds with both tempers present (Figure 4-17), the average wall thickness was / mm greater than both ceramic types with only one temper (Figure 4-18). The average wallthickness for cooking vessels (Figure 4-19) at the Harris Creek site (8VO24), regardless of temper, was a little below the total sample with an average wall thickness of / mm. Measurements of non-cooking vessels at the site (Figure 4-20) showed averages that compared closely to the overall average of Tick Island vessels at about / mm. Fiber-tempered cooking vessels (Figure 4-21) showed measurements that were above averages for the overall cooking vessels at Tick Island as well as for the overall sample, with an average of / mm. Unsooted fiber-tempered sherds (Figure 4-22) averaged mm +/- 2.1 mm, which was not significantly above that of fiber-tempered cooking vessels at Tick Island. It was above the average overall thickness of non-cooking vessels at the site as well as the overall wall- 43

44 thickness for the entire sample. Dual-tempered cooking sherds (Figure 4-23) had an overall wallthickness average of /- 2.7 mm, and unsooted dual-tempered vessel sherds (Figure 4-24) averaged / mm. Vessel sherds of this temper type seemed the thickest of all regardless of function. Within the type, dual-tempered sherds were thinner if used for cooking. Sponge-tempered cooking vessels (Figure 4-25) had an average wall thickness of / mm, which was thinner than the fiber-tempered and dual-tempered cooking vessels, well within the average of overall cooking vessels at the site, and narrower than the overall average thickness of the entire sample of vessel sherds. The average wall thickness of unsooted sponge-tempered vessel sherds (Figure 4-26) was /- 2.5 mm, which was, surprisingly, slightly below the average thickness of sponge-tempered cooking vessel sherds. This average was well below the average for fiber-tempered and dual-tempered non-cooking vessels, and below the average of the entire sample of non-cooking vessels from the site. It was also a little below the overall average of vessel wall thickness from the entire site. Summary of Results New assays derived from soot deposits in this analysis yielded dates for sponge-tempered pottery that ranged from as early as /- 40 B.P. to as late as /- 40 B.P. This places purely sponge-tempered ceramics, also known as St. Johns pottery, into a much earlier chronological location when compared with current models of culture history. This is significant because the earlier dates obtained, place purely sponge-tempered pottery into the Late Archaic a period once thought to have only fiber-tempered ceramics, and sponge tempering as a newly acquired unilineal development. Stylistic tendencies among the vessel sherds analyzed co-varied very well with attributes associated with overall paste this being temper. As far as application of designs was 44

45 concerned, purely fiber-tempered vessel sherds had no examples of designs interpreted to be hastily applied. Dual tempered vessels were similar in their design application in that an overwhelming majority of them were also carefully applied. On the other hand, vessels purely tempered with sponge spicules had a majority of haphazardly applied designs. The co-variance continues to a lesser extent with complex and simple elements. There was far more usage of complex elements with fiber-tempered and dual-tempered pottery than there was with sponge tempered. Sponge-tempered vessels were fairly evenly divided between those with complex elements and those with only simple ones, while the other two paste types had mostly complex design elements incised onto their exterior. Of particular significance is the application of a concluding epilogue band below the motif of many vessel sherds. This trait only occurred with sponge-tempered vessels it was nowhere to be found on either fiber-tempered or dualtempered vessel sherds. Conversely, a very small percentage of the sponge-tempered vessels had banding that preceded the overall vessel motif, and they ranged from only one to two bands. There was a good deal of prologue banding on fiber-tempered vessel sherds ranging from 3-5 bands and an outlying sherd with ten bands on it. Similarly, the dual-tempered vessels exhibited a good deal of preceding bands and ranged from one to four bands, with one outlying sherd having eight bands preceding the design motif. When looking at attributes that would be considered functional, it was found that there was a much smaller proportion of purely fiber-tempered vessel sherds than was originally thought from microscopic inspection. There were only 19 vessels found without sponge spicules in the paste alongside fiber when viewed microscopically. There was a greater representation of dual-tempered vessels, and a slight majority of vessel sherds tempered only with sponge. As far as cooking function is concerned, the vast majority of fiber-tempered vessels were 45

46 unsooted, as was the same with the dual-tempered vessels. Exhibiting opposite tendencies were the sponge-tempered vessels with the great majority of vessel sherds having soot deposits on the exterior surface. The form of these vessels varied according to paste and temper as well. Vessels described earlier as boat-shaped vessels were exclusive to sponge-tempered sherds. The majority of fibertempered and dual-tempered vessels were flat walled in form with the minority being bowls. Sponge-tempered vessel sherds, on the other hand, were by and large bowl shaped in form with a minority of flat-walled vessels, and a small representation of the aforementioned boat shape. As far as form and function is concerned among the fiber-tempered vessels, the majority of sherds used over an open flame were bowls with a minor representation of flat-walled vessels. It is important to note, though, that there were only five fiber-tempered vessels with soot deposit, so this set of statistics can not make a significant statement concerning this particular form and temper as far as function is concerned. Dual-tempered sherds had more sooted flat-walled specimens than bowls, as well as more unsooted flat-walled sherds than bowls, and they were exactly divided between sooted and unsooted bowls. There were a greater number of unsooted flat-walled sherds than there was sooted among the dual-tempered vessel sherds. Vessel sherds tempered with sponge had the majority of bowl forms being sooted, and the majority of flat-wall forms unsooted. When looking at wall thickness, the fiber-tempered vessels were found to be thicker than average, and the dual-tempered vessels thicker than that. Sponge-tempered vessels, on the other hand, were thinner than the entire sample's average. All vessel sherds found to be associated with cooking were also thinner than the overall average, while those that were not were thicker. The small number of fiber-tempered cooking vessels were thicker than the average, as well as the 46

47 substantially larger population of unsooted ones. The dual-tempered cooking vessel sherds were also thicker than the overall average, as well as being thicker than the population of fibertempered cooking vessels. This statement also holds true with the unsooted dual-tempered vessel sherds they were thicker than the overall average and the average of unsooted fiber-tempered vessels. Contrasting this were the sponge-tempered cooking vessel sherds which were thinner than the overall average of vessels, but, surprisingly, the non-cooking sponge-tempered sherds were slightly thinner on average than those with soot deposits. The trends of this analysis point to overall similarities between vessels with dual temper and fiber temper, but contrast starkly with vessels tempered only with sponge spicules. Stylistic and functional traits all co-vary rather nicely between those with fiber and dual tempering, and those that are tempered with sponge. This leads one to believe that style and techno-function are not mutually exclusive categories, as there are some many interrelated traits that co-vary depending on temper, including stylistic application, vessel form, and cooking function (included in this category are soot-deposits and wall thickness). Style could very well be synonymous with function. 47

48 Table 4-2. AMS Dates of St. Johns Ceramics from the Harris Creek Site (8VO24) on Tick Island Lab Number & Site Sample 2 Sigma Calibration 13C/12C Ratio Beta Sample: 8VO24-A348.3 Beta Sample: 8VO24-A Beta Sample: 8VO24-A349.7 Beta Sample: 8VO (Cal BC) and (Cal BP) and Measured Radio Carbon Age (BP) / / / / / / / /00 Conventional Radiocarbon Age (BP) / / / /- 40 Beta Sample: 8VO24-A / / /- 40 Analysis:AMS-Standard delivery. Material: soot. Pretreatment: acid/alkali/acid Table 4-3. Vessel Type and the Care Taken in Design Application Vessel Frequency Fiber Tempered Sponge Tempered Dual Tempered Total Hasty Application Careful Application Total Row Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Hasty Application Careful Application Total Column Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Hasty Application Careful Application Total

49 Table 4-4. Vessel Type and the Complexity of Elements Applied to Vessel Vessel Frequency Fiber Tempered Sponge Tempered Dual Tempered Total Simple Elements Alone Complex Elements Total Row Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Simple Elements Alone Complex Elements Total Column Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Simple Elements Alone Complex Elements Total Table 4-5. The Location of Rim Bands on Different Vessel Types Vessel Frequency Fiber Tempered Sponge Tempered Dual Tempered Total Prologue Lines Epilogue Lines Total Row Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Prologue Lines Epilogue Lines Total Column Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Prologue Lines Epilogue Lines Total

50 Table 4-6. Type and Function as Evident by Soot Deposits Vessel Frequency Fiber Tempered Sponge Tempered Dual Tempered Total Soot Present Soot Absent Total Row Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Soot Present Soot Absent Total Column Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Soot Present Soot Absent Total Table 4-7. Form and Vessel Temper Vessel Frequency Fiber Tempered Sponge Tempered Dual Tempered Total Boat Shaped Bowl Shaped Flat Walled Total Row Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Boat Shaped Bowl Shaped Flat Walled Total Column Percentages Fiber Tempered Sponge Tempered Dual Tempered Total Boat Shaped Bowl Shaped Flat Walled Total

51 Table 4-8. Form and Temper of Vessel as it Related to Function Vessel Frequency FTST FTnoST STST STnoST DTST DTnoST Total Boat Shaped Bowl Shaped Flat Walled Total Row Percentages FTST FTnoST STST STnoST DTST DTnoST Total Boat Shaped Bowl Shaped Flat Walled Total Column Percentages FTST FTnoST STST STnoST DTST DTnoST Total Boat Shaped Bowl Shaped Flat Walled Total Figure 4-7. Profile Form and Pattern Design for Vessel #A Figure 4-8. Vessel Profile of #A with Exterior Pattern Design 51

52 Figure 4-9. Profile of Vessel A From Tick Island. Figure Pattern design found on AMS dated vessel # Figure Incised Surface Design on Vessel # Occurences of Rim Bands Above Vessel Design on Fiber-tempered Pottery Frequency Number of Prologue Rims 10 Figure Frequency of occurrence of preceding band counts on fiber-tempered pottery. Min: 3, Max: 10, Av: 4.8 lines, STD: +/

53 Occurences of Rim Bands Above Vessel Design on Sponge-tempered Pottery Frequency Frequency Number of of Prologue Bands Figure Frequency of occurrence of preceding band counts on sponge-tempered pottery. Min: 1, Max: 2, Av: 1.8 lines, STD: +/-.4. Frequency Number of Prologue Bands Figure Frequency of occurrence of preceding band counts on dual-tempered pottery. Min: 1, Max: 8, Av: 2.35 lines, STD: +/

54 Wall Thickness of Vessel Sherds From the Harris Creek Site (8VO24) 20 Frequency (mm) (mm) Figure Wall thickness of Harris Creek (8VO24) vessel sherds. Max: 19 mm, Min: 2.5 mm, Avg: 9.37 mm, STD: +/ mm. Wall Thickness of Fiber-tempered Vessel Sherds Frequency (mm) Figure Wall thickness frequency for fiber-tempered vessel sherds at the Harris Creek Site (8VO24). Max: 14 mm, Min: 7 mm, Avg: mm, STD: +/ mm. 54

55 Wall Thickness of Dual-tempered Vessel Sherds Frequency (mm) (mm) Figure Wall thickness frequency for dual-tempered vessel sherds at the Harris Creek Site (8VO24). Max: 19 mm, Min: 5.25 mm, Avg: mm, STD: +/ mm Wall Thickness of Sponge-tempered Vessel Sherds Frequency (mm) (mm) Figure Wall thickness frequency for sponge-tempered vessel sherds at the Harris Creek Site (8VO24). Max: 16 mm, Min: 2.5 mm, Avg: 8.22 mm, STD: +/- 2.4 mm. 55

56 Wall Thickness of Cooking Vessel Sherds Frequency (mm) Figure Wall thickness frequency of cooking vessel sherds at the Harris Creek Site(8VO24) Max: 18 mm, Min: 2.5 mm, Avg: 8.95 mm STD: +/ mm. Wall Thickness of Un-sooted Vessel Sherds Frequency Frequency (mm) Figure Wall thickness frequency of unsooted vessel sherds at the Harris Creek Site (8VO24) Max: 19 mm, Min: 4.25 mm, Avg: 9.84 mm STD: +/ mm. 56

57 Wall Thickness of Fiber-tempered Cooking Vessel Sherds Frequency Frequency (mm) Figure Wall thickness frequency of sooted fiber-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 12 mm, Min: 8 mm, Avg: mm STD: +/ mm. Wall Thickness of Un-sooted Fiber-tempered Vessel Sherds Frequency (mm) (mm) Figure Wall thickness frequency of unsooted fiber-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 14 mm, Min: 7 mm, Avg: mm STD: +/- 2.1 mm. 57

58 Wall Thickness of Dual-tempered Cooking Vessel Sherds Frequency (mm) Figure Wall thickness frequency of sooted dual-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 18 mm, Min: 5.25 mm, Avg: mm STD: +/- 2.7 mm. Wall Thickness of Un-sooted Dual-tempered Vessel Sherds Frequency (mm) Figure Wall thickness frequency of unsooted dual-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 19 mm, Min: 6.5 mm, Avg: mm STD: +/ mm. 58

59 Wall Thickness of Sponge-tempered Cooking Vessel Sherds 10 Frequency (mm) Figure Wall thickness frequency of sooted sponge-tempered vessel sherds at the Harris Creek Site (8VO24) Max: 16 mm, Min: 2.5 mm, Avg: 8.31 mm STD: +/ mm. Wall Thickness of Un-sooted Sponge Tempered Vessel Sherds 5 Frequency (mm) Figure Wall thickness frequency of unsooted sponge-tempered vessel sherds at the Harris Creek Site (8VO24) Max: mm, Min: 4.25 mm, Avg: 8.08 mm STD: +/- 2.5 mm. 59

60 CHAPTER 5 DISCUSSION AND CONCLUDING STATEMENTS It is obvious from recent radiocarbon dating of sponge-tempered pottery at the Harris Creek Site (8VO24) on Tick Island that the relationship between different forms of tempering needs to be reexamined in a different light. At Tick Island, the new AMS dates clearly show a contemporaneous relationship between sponge-tempered, dual-tempered, and fiber-tempered pottery of the Late Archaic. It is important to note that this is a site-specific study, and its interpretations, therefore, should be applied only to analytical work done on the ceramic specimens from Tick Island and tested with other nearby sites. To extrapolate these findings any further would be inconsistent with one of the main premises of this paper; that is, that data from too few sites have been used to make gross generalizations about the behaviors and materials of southeastern peoples, particularly those residing in Florida. Only when more site specific analyses are done and dates are retrieved from a wider variety of sites, can there be a more detailed and relevant regional analysis, comparing sites throughout the St. Johns River Valley and beyond, in order to provide a better basis for broader generalizations. These new AMS dates for the sponge-tempered vessels at Tick Island clearly indicate that some are contemporaneous with their counterparts of fiber and dual tempering. Questions, therefore, must be answered concerning their relationship in this light. That was the focus of this analysis; to chart similarities and differences between these vessels as to their form, function, and design, and to propose answers to questions concerning why these two separate vessel types were used by the same peoples. It is the intent of this discussion to frame an explanatory model that takes into account the social and political context in which these vessels were being produced and utilized by the inhabitants of Tick Island. 60

61 Discussion of Results Style: The results clearly show both distinct differences and similarities in the application of incised surface designs to both vessel types. The similarities lie in the complex and simple design elements used to produce vessel motifs. They were created using the same design elements, but the differences were defined in the application of these designs. Fiber-tempered and dual-tempered vessel sherds were shown to have a carefully applied design with a far greater number of complex elements, while sponge-tempered vessel sherd designs were shown to have been very haphazardly applied with less complex elements. Differences also lie in the banding of vessels. Fiber-tempered and dual-tempered vessel sherds had a large range of possible band numbers preceding the various vessel motifs ranging from 1-10, while most of the spongetempered vessels had their design concluded at the bottom with one horizontal band. This band usually occurred one-quarter to midway down the vessel with the rest of the surface left blank. Very few sponge-tempered specimens had preceding rim bands, and in the rare instances that they did occur, they were limited to one or two bands. Both vessel types were being made at the same time incorporating the same values inherent in the design elements, yet the degree of care used in application and grammatical approach to applying these designs were distinctly different. It is believed here that this is due to a need to expediently produce certain vessels that were less important in aspects of social/ritual functions. Pots tempered with fiber or both tempers would have served a more ceremonial significance than those tempered with sponge. This is supported by the technofunctional analysis conducted with this collection. Techno-function: The tempering of what Bullen would consider St. Johns and Orange vessels proved to be as expected, based on the premise of their original typology, that is, Orange specimens being tempered with fiber inclusions, and St. Johns being tempered with sponge- 61

62 spicules. What was unexpected, however, was the number of what would be typed as Orange vessels that were dual-tempered containing both a fiber and sponge-spicule paste enhancing previous analysis done by Cordell (2004). There were more dual-tempered Orange vessels than there were purely fibered temper vessels. Out of 69 Orange vessels, only 19 contained fiber alone, while the remaining 50 had both sponge and fiber inclusions in the paste. As far as function is concerned, it was shown that there were many more spongetempered sherds used over an open flame then there were from the fiber-tempered and dualtempered sample. It can be reasonably inferred from this that sponge-tempered vessels were more likely to be used in cooking since soot deposits are fairly clear evidence of a cooking function. When form and function are considered in comparing specimens from the temper categories, there was a much larger proportion of flat-walled vessels than there were bowl shapes among the fiber-tempered and dual-tempered specimens, and conversely, there was a preponderance of bowl shapes over flat-walled vessels among the sponge-tempered specimens. When looking at these forms as a matter of function, fiber-tempered and dual-tempered bowls were evenly divided between those with soot, and those without. Analysis of sponge-tempered bowls, however, showed that there were a far greater number of these vessel types that were used over an open flame than were not. When looking at the flat-walled vessels represented in this study, fiber-tempered and dual-tempered vessels were shown to be used to a much greater extent for non-cooking functions as evidenced by the lack of soot deposits on most of these vessels. Conversely, specimens of sponge-tempered flat-walled vessel sherds included a much larger proportion that were used over an open flame compared to a much smaller number that were not. Sponge-tempered boat-shaped vessels were fairly evenly divided between those with exterior soot deposits present and those lacking it. 62

63 Measurements of wall-thickness as an attribute showed fiber-tempered and dualtempered unsooted vessels to be thicker than those that had exterior soot deposits, indicating that cooking vessels were thinner than those not used for cooking. Fiber-tempered and dual-tempered vessel sherds, on average, also tended to be considerably thicker than the sponge-tempered vessels in the total sample. Social Implications of Ceramic Analytical Results from Tick Island It is inferred from the results of this analysis that the following picture was occurring at the Harris Creek Site (8VO24) on Tick Island. The site was a place of ceremonial importance. Of this there is little doubt. Also believed here is that mortuary feasts of a very ceremonial nature were an important aspect of this site, and its location had a definite spiritual and social significance to those that utilized it. This belief is based on the evidence that this site was a place of burial utilized in the pre-ceramic archaic period (Aten 1999), and it is assumed that the continued utilization of this site in conjunction with these burials would have given it great spiritual importance. The period between the Middle and Late Archaic saw a sharp rise in population, increasing the participants in these feasts. With that increase in population and ritual participation developed an increase in the demands on labor among potters, which are assumed here to be women. It is further believed that the ceremonial and ritual serving aspects were reserved for the fiber-tempered and dual-tempered ceramics, whose techno-functional characteristics make them less conducive to cooking and more conducive to serving and storing food. There were some sooted examples of both. However, the majority of these temper types were lacking in soot. The sponge-tempered vessels were much thinner and more effective at heat transfer from the exterior, making them more useful cooking tools. Also, there is a higher degree of soot present on the 63

64 sponge-tempered vessel sherds than on the others, lending further credence to the idea that they were used over an open flame. Within the fiber-tempered and dual-tempered population there was a much higher proportion of flat-walled vessels, a form also more likely to imply a serving function as food can be served readily in flat-walled and flat-bottomed vessels (similar to casserole dishes). The higher number of bowl-shaped pots within the St. Johns sample would have been better for suspending and cooking food-stuffs over an open flame. A rounded vessel is much better for stirring the contents within than a flat-walled and flat-bottomed vessel in which the heated contents could have been transferred at the ceremonial feasts. It is theorized here from the data that the sponge-tempered vessels served a more functional or secular purpose at Tick Island mortuary feasts, while contemporary fiber-tempered and dual-tempered vessels served a ceremonial or sacred function at these events. Fiber-tempered and dual-tempered vessel sherds had much more complex and elaborate designs incised onto their exterior covering the entire vessel, but most of the sponge-tempered samples had very poorly executed designs that did not cover most of the vessel, being abruptly halted midway down the vessel by a concluding band. It is posited here that the rapidly expanding population in the Late Archaic necessitated an increase in production of the more functional vessels at the site to cook for a larger number of people. This could explain the hastily applied designs on these vessels as they were being produced very rapidly, but there was still an importance in having these design elements and motifs on the pots, otherwise they would have been abandoned altogether. Conclusion It has been demonstrated through this study that broad generalizations cannot be made about the St. Johns River valley and neighboring areas during the Late Archaic period given the limited data and dates that are available concerning these periods. A minimal number of dates 64

65 have been used to make generalizations and form a cultural-chronological sequence based on ceramic typology that has been shown through new AMS dates to need serious revision and rethinking. There is a need for further qualitative studies of specific sites a for the collection of as many dates as possible in order to address the many differences found in the nature and rate of cultural development at various sites. Even the stylistic grammar and use of the same design elements might differ at various locations along the river. Only when more qualitative studies are done of individual sites that good, relevant regional analyses can be done, and more broad generalizations made As far as style is concerned, this paper has not attempted an in-depth analysis of the grammatical rules inherent in design application on these pots. It does, however, offer an insight into differences in application and complexity of designs between three different yet contemporaneous temper-types of pottery existing at the same site, and used (most likely) by the same people. It is the intent here to promote development of an effective approach to understanding the grammatical rules of applying design elements and creating motifs at this site and at others. The important thing is to improve understanding of culture history as forms of social process and ideology. People have always had the ability to utilize and manipulate symbols, and this is an area of key interest to anthropologists. It devalues the heritage of the native population of this continent to relegate their history to the natural sciences under the false dichotomy of history versus prehistory. Prehistory has been defined as a period of literacy by the western standards of written language. But literacy can also be defined as the ability to utilize and manipulate symbols actively used within a given culture. Conventional history views the time before the arrival of European influence as a history of people passively reacting to 65

66 environmental circumscription. There is compelling evidence, however, suggesting an alternative perspective involving a wide range of human reactions to environmental change, all rooted in culture and ideology. If there truly is a prehistory, it would be defined not by the subject s illiteracy, but by the illiteracy of the observers in understanding the symbols and their proper grammatical use within that given culture. 66

67 APPENDIX A FIBER-TEMPERED AND DUAL TEMPERED INCISED VESSEL MOTIFS Figure A-1. Examples of fiber-tempered and dual-tempered incised vessel designs 67

68 Figure A-2. Additional examples of fiber-tempered and dual-tempered incised vessel designs 68

69 APPENDIX B SPONGE-TEMPERED INCISED VESSEL MOTIFS Figure B-1. An example of an incised motif applied to a sponge-tempered vessel 69

70 Figure B-2. Additional examples of motifs found on sponge-tempered vessels 70

71 APPENDIX C VESSEL PROFILE FORMS FROM TICK ISLAND Figure C-1. Examples of vessel profiles from the Harris Creek Site (8VO24) on Tick Island 71

72 APPENDIX D PHOTOGRAPHS OF INCISED VESSELS FROM TICK ISLAND Figure D-1. Photographs of dual-tempered and fiber-tempered incised ceramics from the Harris Creek Site (8VO24) on Tick Island. 72

73 Figure D-2. Photographs of sponge-tempered incised ceramics from the Harris Creek Site (8VO24) on Tick Island. 73