ThesisPDF Available

An Exploration of Turtle Shell Rattle Manufacture in the Mississippian Period

Authors:

Abstract and Figures

Please see my newer papers: "An Experimental Study of Turtle Shell Rattle Production and the Implications for Archaeofaunal Assemblages" published by PLOS ONE (https://doi.org/10.1371/journal.pone.0201472) and "Identifying Turtle Shell Rattles in the Archaeological Record of the Southeastern United States" published by Ethnobiology Letters (https://www.researchgate.net/publication/322001494_Identifying_Turtle_Shell_Rattles_in_the_Archaeological_Record_of_the_Southeastern_United_States).
Content may be subject to copyright.
An Exploration of Turtle Shell Rattle Manufacture in the Mississippian Period
Anthropology Senior Thesis
Presented to
the Anthropology Faculty
Andrew D. Brown
In Partial Fulfillment
of the Requirements for the Degree
Bachelor of Science in the
Department of Sociology and Anthropology
Middle Tennessee State University
November 2011
ii
ACKNOWLEDGMENTS
I would like to thank all of the people who have helped me with this research, and with all of my
questions. Special thanks to my professor, mentor, and guide on this research project, Dr. Tanya
Peres, who served on the thesis committee as the director. She met and discussed this research
throughout the process, checked my re-analysis of the turtle remains, was very supportive
throughout the process, and provided substantial editorial comments on the text and organization
of this thesis. Thanks to Dr. Kevin Smith for providing direction for experimental archaeology
and sharing other possibilities for this research, providing different background information,
discussing this research, serving on the thesis committee. Thanks to Dr. Will Leggett and Dr.
Eller for serving on this thesis committee and providing many additional comments and
suggestions on the paper. Thanks to all those who have donated turtle shells Rusty Simmons,
Austin Kreitzer, Rae Smith, Megan O’Connell. Thanks to Rusty Simmons for allowing me to use
his yard to prepare the turtle remains for the experimental archaeology. Thanks to Mark Norton,
Tennessee Division of Archaeology (TDOA), for the loan of his bow and chert drill and an
explanation of how to use it. Thanks to Aaron Deter-Wolf (TDOA) for providing additional sites
and background information, searching databases, and answering questions. Thanks to Suzanne
Hoyal (TDOA) who helped me go through the Penitentiary Branch reports and attempted to fix
some photos that had flood damage. Information about the Warren Wilson site was provided by
R. P. Stephen Davis Jr. (University of North Carolina-Chapel Hill), and the Garden Creek Site
by Alice Wright (University of Michigan). Thanks to Dan Morse, University of Tennessee, Pope
Museum, and Dr. David Hally for providing photos. Thanks to the families in Castalian Springs
for allowing the search for turtles on their property. This research was funded through a MTSU
iii
Undergraduate Research and Creative Activity Grant. I have a deep appreciation for my wife,
Bailey, for supporting, listening, and encouraging me through all of this research and for
editorial comments.
iv
TABLE OF CONTENTS
Acknowledgments.............................................................................................................. iii
List of Figures .................................................................................................................... vi
List of Tables ..................................................................................................................... ix
Section Page
Introduction ....................................................................................................................... 1
Background ....................................................................................................................... 2
Modified and Unmodified Uses of Turtle ....................................................................... 2
Symbolic significance of turtle ..................................................................................... 3
Turtle as a food resource ............................................................................................. 4
Archaeological turtle shell rattles ................................................................................ 4
Turtle shell bowls, cups, and spoons ............................................................................ 7
Expectations of Turtle Shell Rattle Use in the Archaeological Record .......................... 7
Context ......................................................................................................................... 8
Turtle species used for rattles ...................................................................................... 8
Turtle skeletal representation in the archaeological record ........................................ 9
Modifications .............................................................................................................. 10
Residue/staining ......................................................................................................... 11
Rattle implements ....................................................................................................... 11
Turtle Rattle Types ........................................................................................................ 14
Body rattle .................................................................................................................. 15
Handheld rattle .......................................................................................................... 16
Wooden staff rattle ..................................................................................................... 16
Southeastern United States Ethnographic and Ethnohistoric Evidence ........................ 17
Archaeological Evidence from the Mississippian Southeast ......................................... 21
Mitchell, Illinois ......................................................................................................... 21
Zebree, Arkansas ........................................................................................................ 26
Moundville, Alabama ................................................................................................. 22
King, Northwestern Georgia ...................................................................................... 23
Warren Wilson, North Carolina ................................................................................. 24
Coweeta Creek, North Carolina ................................................................................ 26
7HA1 and 8HA1, East Tennessee ............................................................................... 26
DeArmond, East Tennessee ........................................................................................ 27
Hamilton County, Tennessee ...................................................................................... 27
Toqua site, East Tennessee ........................................................................................ 28
Hiwassee Island, East Tennessee ............................................................................... 28
Other possible sites .................................................................................................... 29
v
Discussion of the Archaeological Correlates ................................................................. 29
Location of turtle shell rattles .................................................................................... 29
Proposed rattle characteristics .................................................................................. 31
Age and sex of burials ................................................................................................ 31
Experimental Archaeology and Turtle Shell Rattle Manufacture ............................. 33
Methods ......................................................................................................................... 33
Approach and design .................................................................................................. 33
Specimens ................................................................................................................... 35
Documenting the process ........................................................................................... 36
Turtle Shell Rattle Production ....................................................................................... 37
Drilling instruments ................................................................................................... 37
Drilling the carapace ................................................................................................. 38
Breakage patterns ...................................................................................................... 47
Rattle implements ....................................................................................................... 49
Cordage ...................................................................................................................... 50
Summary ........................................................................................................................ 50
Case Study Sites: Fewkes and Castalian Springs......................................................... 50
Fewkes (40WM1) .......................................................................................................... 51
Fewkes turtle re-analysis ........................................................................................... 52
Castalian Springs (40SU14) .......................................................................................... 59
Castalian Springs turtle re-analysis ........................................................................... 59
Discussion of Fewkes and Castalian Springs ................................................................ 63
Discussion and Summary ............................................................................................... 65
Discussion and Future Research .................................................................................... 65
Summary ........................................................................................................................ 68
References Cited.............................................................................................................. 71
vi
LIST OF FIGURES
Figure Page
Figure 1 Modern Eastern Box Turtle Plastron and Carapace ......................................9
Figure 2 Two types of turtle shell body rattles ..........................................................10
Figure 3 Rattle pebbles from Burial 16 at Warren Wilson ........................................12
Figure 4 Turtle shell leg rattle without drilled holes .................................................15
Figure 5 California wooden staff rattle ......................................................................17
Figure 6 Creek woman’s leg shackles or leggings which are attached to leather......19
Figure 7 Turtle shell remains from Burial 16 at Warren Wilson ...............................22
Figure 8 Turtle shell remains from Burial 29 at Warren Wilson ...............................24
Figure 9 Complete turtle carapace from Burial 29 at Warren Wilson .......................25
Figure 10 Recreation of a Dallas woman, wearing turtle shell rattles on each arm ....25
Figure 11 Burial 223 at King .......................................................................................26
Figure 12 Turtle shell rattle from Zebree .....................................................................28
Figure 13 Turtle shell rattle from Burial 26 at Ensworth ............................................34
Figure 14 Experimental Plastron .................................................................................35
Figure 15 Drilling Turtle Shell with River Cane .........................................................37
Figure 16 Bow and chert drill ......................................................................................38
Figure 17 Bow and chert drill in action .......................................................................38
Figure 18 The beginning of drilling the carapace ........................................................39
Figure 19 Oval shape from drill ...................................................................................39
Figure 20 Circular shape from changing drill angle ....................................................40
Figure 21 Before using river cane to polish out excess bone ......................................40
vii
Figure 22 After using river cane to polish out excess bone .........................................41
Figure 23 Ridge in the central part of the drill opening ...............................................41
Figure 24 Experimental Marginal ................................................................................42
Figure 25 Experimental Marginal after break ..............................................................42
Figure 26 Experimental Costal A ................................................................................43
Figure 27 Experimental Costal B .................................................................................43
Figure 28 Experimental Costal B after the split ..........................................................44
Figure 29 Experimental Costal B with break indentation ............................................44
Figure 30 Experimental Carapace A ............................................................................45
Figure 31 Experimental Carapace B ............................................................................45
Figure 32 Experimental Hypoplastron Central ............................................................46
Figure 33 Breakage patterns: sharp angle ....................................................................48
Figure 34 Drill indentation ..........................................................................................49
Figure 35 River pebbles collected from Stones River .................................................49
Figure 36 Map of Fewkes and Castalian Springs ........................................................52
Figure 37 Emydidae marginal stained with red ochre recovered from
Feature 55 at Fewkes ...............................................................53
Figure 38 Partially drilled box turtle marginal recovered from
Feature 55 at Fewkes ...............................................................54
Figure 39 V-shaped break and cancellous portion of the shell recovered
from Feature 184 at Fewkes .....................................................54
Figure 40 Stained with red ochre and exhibits sharp angles recovered from
Feature 549 at Fewkes .............................................................55
Figure 41 Drilled/Perforation in soft-shelled turtle recovered from
Feature 818 at Fewkes .............................................................56
Figure 42 Worked Turtle Shell recovered from Test Unit 42 at Fewkes ....................57
viii
Figure 43 Ground marginals recovered from Test Unit 100 at Fewkes ......................58
Figure 44 Drilled turtle shell recovered from Test Unit 145 at Fewkes ......................58
Figure 45 Drilled pond-slider turtle recovered from Feature 4
at Castalian Springs..................................................................60
Figure 46 Polished turtle shell recovered from Feature 4 at Castalian Springs ...........60
Figure 47 Breakage patterns: sharp angles recovered from Feature 23
at Castalian Springs..................................................................61
Figure 48 Two complete box turtle carapaces from Feature 119
at Castalian Springs..................................................................62
ix
LIST OF TABLES
Table Page
Table 1 Mississippian Period Archaeological Evidence for Turtle Shell Rattle.......5
Table 2 Archaeological Correlates Rattle Characteristics Checklist ......................13
Table 3 Sex Distribution of Archaeological Burials ...............................................32
Table 4 Age Distribution of Archaeological Burials ..............................................32
Table 5 Final Measurements of the Ten Experimental Drill Openings ..................47
Table 6 Fewkes Associated Rattle Modifications (Features Only).........................53
Table 7 Fewkes Associated Rattle Modifications (All, including Features) ..........56
Table 8 Castalian Springs Associated Rattle Modifications ...................................59
1
Introduction
Turtle remains are recovered frequently from archaeological sites in the Eastern United
States spanning the Paleoindian to Historic periods. The most commonly found species in the
southeastern United States is the eastern box turtle (Terrapene carolina). Interpretations of turtle
use ranges from subsistence to ceremonial. The turtle remains recovered in the record sometimes
have signs of modification. In some instances, these modifications could point to the use of turtle
rattles. The present study focuses specifically on the use of turtle shell rattles in the Southeastern
United States during the Mississippian Period (A.D. 850 to A.D. 1700)1 (Walthall 1990).
In several archaeological site reports and publications, complete or partial turtle shells are
deemed rattles while smaller fragments are considered food refuse. Thus, the problem is that the
identification lacked specific indicators and characteristics which would have allowed for a
clearer interpretation on the function of the remains. In order to solve the problem, I propose six
indicators and characteristics that can be used in identifying and interpreting turtle remains,
whether complete, partial, or fragmentary. The six indicators are: context, turtle species, turtle
skeletal representation, modifications, residue/staining, and rattle implements. The indicators as
well as background information2 on construction, materials, common species, and modification
on turtle rattles are derived from ethnographic and ethnohistoric, archaeological, and
experimental archaeology data. Comparative information was also gained from modern
ethnographic studies on the Seneca, Iroquois, Shawnee, Seminole, Cherokee, and Creek. Finally,
1 Around the late A.D. 1700 date, Walthall notes that the Mississippian culture had almost disappeared by this time.
2 I conducted the background research and data collection under the MTSU Undergraduate Research and Creative
Activity Grant.
2
the indicators were tested against two Mississippian Period sites in Middle Tennessee, Fewkes
and Castalian Springs.
The experimental archaeology I conducted made clearer the archaeological turtle shell
rattles and the process of construction. I thoroughly documented the construction of the
experimental rattle3 which proved to be very important for the identification of turtle rattles at
the case study sites. I performed a deeper analysis on the primary data collected from these two
sites in order to explore the modification of turtle remains in relation to turtle shell rattles.
Background
Modified and Unmodified Uses of Turtle
Archaeological interpretations of turtle use include: food, adornment, rattles, cups, bowls,
spoons, medicine bundles4 and bags, effigies, and sacrificial purposes. At the most basic level,
archaeological turtle remains can be separated into two categories: modified and unmodified.
Modified remains can be further divided into two sub-categories: intentional and unintentional
modification. Intentional modification is a planned alteration to a specimen by humans. An
unintentional modification is an unplanned alteration to a specimen by humans. Intentional and
unintentional modification can be further divided into two sub-categories: non-subsistence and
subsistence. Non-subsistence refers to the use of a specimen for purposes other than food, such
as rattles that have been intentionally modified or turtle containers that have been used to hold a
non-food material. Subsistence refers to the use of a specimen as a food-related resource.
3 The experimental rattle is the complete turtle shell rattle that I constructed through experimental archaeology.
4 Kevin Smith, personal communication 2011.
3
Subsistence can be further divided into two sub-categories: dinnerware and consumption. The
term, ―dinnerware‖ refers to the use of a specimen for holding or serving food. Consumption is
the use of specimen for dietary purposes. Unmodified remains can be further divided into two
sub-categories: non-subsistence and subsistence. For unmodified remains, non-subsistence
coincides with the definition under modified remains; however, it refers to rattles that have not
been intentionally modified.
Symbolic significance of turtle. Objects gain meaning through an interaction with human
culture. Then, objects can become symbols. When a human collects a raw natural material, the
material is brought into a cultural dialogue. The modified or un-modified object gains additional
meaning from the cultural conversation and interaction. For the archaeological raw materials, the
context of the object can ascribe a genre of meaning, such as ritual or ceremonial (Kelly 2006).
For the Cherokee, daksi, box turtles, are revered for their strong armor and ability to have
a strong hold in their fortress against impending threats. The terrestrial-dwelling box turtle is
looked upon as a tough soldier in the Middle World (Fradkin 1990). Among the Cherokee
ideology, the box turtle transforms into a great healer, who is able to cure aliments and
rheumatism (Fradkin 1990). The box turtle has the ability to be on land or water (Pechanga
2010). For the Shawnee, Iroquois, and other Woodland tribes (including the Algonkian), the
world was placed on top of the turtle’s shell (Reilly and Garber 2007). For Mississippian pre-
warfare rituals, a space was set up away from the village in order to separate the warriors from
the general public. This was regarded as a sacred space, which was used for the cleansing and
preparation of the participants (Reilly and Garber 2007:157). This suggests that there is a belief
in the transformation of the ordinary into the sacred, which could possibly be applied to turtle in
4
regards to subsistence and/or non-subsistence use. The different views and uses of turtle suggest
a symbolic significance (Trevelyan 2004).
Turtle as a food resource. Fragmentary turtle remains are often considered food refuse. It is not
clear whether the Mississippian people using turtle as a food resource. Archie Carr (1952) gives
ecological details and the life history of several species and sub-species of the eastern box turtle.
Box turtles can eat poisonous mushrooms with impunity which may make them poisonous to
humans. He explains a few instances or stories of people eating box turtle then becoming ill,
although no serious injury has been documented (Carr 1952:147). Other authors such as Dodd
(2002:114) report that they are ―quite poisonous‖ to humans. Babcock also tells a story of miners
in New England getting sick from eating box turtles (Babcock 1919). There has been no known
scientific study of the effects of eating box turtle; thus the actual toxicity of box turtles needs to
be tested. Turtle elements, excluding the carapace and plastron, should be investigated in the
archaeological record in order to see if they occur in food refuse areas. An analysis of food
refuse areas could reveal information about whether turtles were part of the diet. Investigations
should be conducted on the areas that only retain carapace and plastron fragments in order to see
if the contexts differ from the other turtle elements.
Archaeological turtle shell rattles. Turtle shell rattles have been recovered from numerous
archaeological contexts and time periods. One of the oldest recoveries was the Paleoindian
Period Horn Shelter site dating to around 11,000 BP (Texas Archaeological Research Laboratory
2010). Numerous archaic sites yield rattles such as Penitentiary Branch (Cridlebaugh 1986) and
Ensworth (40DV184) (Deter-Wolf et al. 2004). Woodland sites yield rattles such as Apple Creek
(Parmalee et al. 1972), Little Turkey Hill, and Butler County, Missouri (Morse and Morse 2009).
5
Western Woodland sites include seventh century AD at Vandal Cave, 600-1000 AD at Pecos
(Brown 1967). Western Mississippian sites and groups include the Mogollon and Anasazi
(Brown 1967), and the Oro Grande Mojave (Langenwalter 1983). The Southeastern
Mississippian Period sites have a rich archaeological rattle record including Mitchell (Sampson
and Sampson and Esarey 1993), Zebree (Morse and Morse 2009), Moundville (Wilson and
Davis 2003; Wilson et al. 2006), King (Hally 2009), Warren Wilson (Rodning and Moore 2010),
Coweeta (Rodning and Moore 2010), Dallas sites (7HA1 and 8HA1) (Lewis and Lewis 1995),
DeArmond (McCarthy 2011), Hamilton County, Tennessee (Lewis 1943), and Toqua (Chapman
1995) (Table 1). Ethnohistoric sites include Caplen Mound (Texas Archaeological Research
Laboratory 2008) and Dutch Hollow (Carpenter 1942). Kenneth Orr (1946) associates caches of
small pebbles with turtle rattles at Spiro. He also associates box turtle shells with turtle rattles at
the Littlefield site. Some constructions are still made in the present day by groups like the
Pechanga Band (Pechanga 2010), the Creek, and Seminoles (White Deer 1995). Historic and
modern samples such as the Seneca, Iroquois, Shawnee, Seminole, Cherokee, and Creek that are
considered to be post-historic will be used for comparative purposes.
Table 1. Mississippian Period Archaeological Evidence for Turtle Shell Rattles.
Mississippian Southeast
Site
Time Period
Context
Additional Information
Mitchell,
Illinois
1150-1250
CE
Substitute rattle: Copper
Turtle Rattles
Zebree,
Arkansas
1350-1650
AD
House floor
Moundville,
Alabama
Early
Mississippian
Ritual multiple
household area
6
Table 1 (continued). Mississippian Period Archaeological Evidence for Turtle Shell Rattles.
Site
Time Period
Context
Additional Information
King,
Georgia
1300-1500 AD
2 Burials
(1female and 1
possible female)
Warren
Wilson,
North
Carolina
Dallas Phase
Mississippian
ca. 1200-1500
AD
2 Burials (1 male
and 1 female)
Coweeta
Creek, North
Carolina
1400-1700 AD
2 Burials
(female)
Dallas Site
(8HA1), East
Tennessee
Dallas Phase
1 Burial
(possible female)
DeArmond
AD 1200-1450
2 Burials
(female)
Hamilton
county, East
Tennessee
1400-1500 AD
1 Burial (female)
Part of the Dallas group
Toqua, East
Tennessee
Late
Mississippian
period
1 Burial (female)
Dallas Phase
Hiwassee
Island, East
Tennessee
Dallas Focus -
1200-1500 AD
3 burials (2
female and 1
male)
Spiro,
Oklahoma?
800-1450 AD
Caches of small pebbles
= turtle rattles
Littlefield
site,
Oklahoma?
800-1450 AD
4 Burials
Box turtle shells = turtle
rattles
Mound
Bottom,
Tennessee?
900-1300 AD
Turtle shells used as
rattles
Garden
Creek, North
Carolina?
Male burial?
7
Turtle shell bowls, cups, and spoons. Turtles have been used for several purposes in addition to
consumption. These purposes are discussed in order to determine identification and function of
turtle artifacts in order to distinguish them from turtle shell rattles. Turtle shell bowls, cups, and
spoons are noted at several sites including: Sunwatch (Cook 2007), Sponemann (Fortier et al.
1992), Bonham (Boyd et al. 2005), Apple Creek (Parmalee et al. 1972), and Windover (Penders
2002). In the late Archaic Period (4000-1000 BC), inland peoples such as those at Lake Lamoka
in New York, used turtle cups, which were probably associated with ceremonies or for curing
disease (Griffin 1967:180). These items are not necessarily food-related but could be for
ceremonial, ritual (Fradkin 1990), medicinal, or other daily uses, and are usually associated with
burials (Lewis and Lewis 1995). During the late Mississippian Period (AD 1350-1650) (Morse
and Morse 2009), two types of turtle effigies, turtle bone and ceramic show up in the Central
Mississippi Valley, which is located West of the Mississippi River and spans from about 25
miles north of Memphis, Tennessee to about 30 miles north of New Madrid, Missouri (Morse
and Morse 2009). For turtle bone, a turtle shell was carved into a turtle effigy. For ceramic, two
items were made to look like turtle including a pottery turtle effigy and a turtle effigy tetrapod
bottle.
Expectations of Turtle Shell Rattle Use in the Archaeological Record
The proposed indicators and characteristics for identifying turtle shell rattles in the
archaeological record are context, turtle species, turtle skeletal representation, modifications,
residue/staining, and rattle implements. Each indicator is discussed here and applied to two
archaeological case studies, below.
8
Context. Within the archaeological record, the majority of rattles or possible rattles are recovered
from ceremonial and burial contexts in Eastern Tennessee. The ceremonial and burial context is
evident for the various Southeastern Mississippian rattles, which will be discussed in-depth in the
archaeological evidence section. Artifactual context is comprised of the physical location of the
artifact and the formation of the surrounding area. The physical location is an arbitrary location
with defined parameters such as strata, level, and unit. The physical location includes a burial
context. A burial context is the removal or addition of soil, for the purpose of concealing human,
faunal, or other remains. Ceremonial contexts are areas that are used for ritual and religious
purposes. A structure that is empty of material objects often suggests that it is ceremonial. The
presence of ritual artifacts or specific structure types within such a context strengthens a
ceremonial interpretation. The second consideration of context is the formation of the area
around the artifact. The formation includes the preparation of an area, addition and removal of
the soil, and post-depositional disturbances.
Turtle species used for rattles. According to ethnohistoric, ethnographic, and archaeological
data, there are several turtle species used for rattles, including: tortoise (Testudinidae), desert
tortoise (Gopherus agassizii), eastern box turtle, snapping turtle (Chelydra serpentina), pond
slider (Trachemys scripta), and Western pond turtle (Clemmys marmorata). However, the three
main species that were used ethnographically and archaeologically in the making of rattles
include the eastern box turtle, the desert tortoise, and the snapping turtle. The native eastern box
turtle was the most commonly used taxa in the Southeast. No information has been presented in
literature on whether turtles were killed specifically for the purpose of rattle manufacture or were
collected already dead. Several shells of eastern box turtle were used for the constructed rattle in
9
my experiment. The Seneca are documented as using snapping turtles for rattles (Parker 1909).
Snapping turtles are also used by the modern Iroquois (Speck 1945). The tortoise and desert
tortoise, while not native to or recovered from archaeological contexts in the Southeastern US,
are incorporated in this section as they represent different rattle types. The desert tortoise is
common in the archaeological record of Southwest US (Schneider 1996). The Oro Grande site,
located just north of Victorville, California, yielded several tortoise remains including the desert
tortoise (Schneider and Everson 1989).
Turtle skeletal representation in the archaeological record. The most common elements for
turtle shell rattles are from the carapace, which include marginals, neurals, and costals, and the
plastron (Figure 1). In some instances, the marginals may be missing, as was the case at
Hiwassee Island (Lewis and Kneberg 1970) (Figure 2). Only eastern box turtle carapaces and/or
plastrons are associated with turtle shell rattles.5 For the snapping turtle, the head is sometimes
associated with the rattle as well.
Figure 1. Modern eastern box turtle plastron and carapace, Left to Right: Hyoplastron,
Hypoplastron, and Carapace (Middle Tennessee State University Zooarchaeology Comparative
Collection).
5 There is the possibility that groups were able to dry out the turtle while leaving the limbs intact (Kevin Smith,
personal communication 2011); however, this hypothesis will need to be tested.
10
Figure 2. Two types of turtle shell body rattles (Lewis and Kneberg 1970: Figure 28).
Modifications. There are two types of modification that will be discussed: intentional and
unintentional modification. Intentional modification is a planned alteration to a specimen in
order to achieve a desired outcome. Intentional modification includes: drilling, cut marks,
grinding/smoothing, scoring and snapping, working, burning, and incising. Unintentional
modification includes: stress striations, wear from rattle objects, thermal alteration6, and polish
caused by use. The most common rattle modifications are drilling, cut marks, polishing, and
6 Burned remains can be placed into both categories because it is difficult to ascertain intentionality.
11
grinding/smoothing. Intentional and unintentional modifications were tested by conducting
experimental archaeology. For experimental archaeology, intentional modifications for drilling
and other modifications were tested and unintentional included striations, wear, and polish.
Residue/staining. Residue from pine pitch and staining from red ochre may be present. This idea
is derived from the fact that asphaltum is used on turtle shell rattles in the Southwest. Asphaltum
was not included in this study because it is a Southwestern US material (Langenwalter 1983;
Schneider 1996). Pine pitch is a glue-like adhesive that is made from raw pine pitch, fine ground
charcoal, and plant materials (Fontaine 2011; Gibby 1999). Pine resin and pitch are materials that
can be used as waterproofing agents as well as glue to mend broken pottery and arrows (Fox et
al. 1995). It could easily be used to fill in gaps, seal off a plastron, or cement the carapace and
plastron back together (see Figure 1).
Rattle implements. A rattle implement is any object that can be inserted into the rattle for the
purpose of sound. Pebbles (Rodning and Moore 2010; University of Michigan 1970), hard seeds
(Capron 1953), and drum teeth (Lewis and Kneberg 1970; Lewis and Lewis 1995) are employed
in order to create the rattle effect. Pebbles are the most recorded in the archaeological recorded
(Figure 3; Table 2), though they are often discarded as fill or backdirt.
12
Figure 3. Rattle pebbles from Burial 16 at Warren Wilson (Photo Courtesy of the Research
Laboratories of Archaeology, The University of North Carolina at Chapel Hill).
The hard seeds (such as Canna sp., wild canna, for the Seminoles), possibly, do not survive in the
archaeological record because the environment is not conducive to the preservation of these
materials. The pharyngeal teeth of the freshwater drum (Aplodinotus grunniens) were used at
Dallas Phase sites (Lewis and Kneberg 1970). In the experimental archeology, river pebbles were
used to create the rattle effect in the rattle.
13
Table 2. Archaeological Correlates Rattle Characteristics Checklist.
Burial
Number
Ceremonial
or Burial
Context
Eastern
Box
Turtle
Only
Carapace
or
Plastron
Modification
Residue
or
Staining
Rattle
Implements
Attached
to arm or
ankle
Mitchell,
Illinois
X
X
Zebree,
Arkansas
X
X
X
X
Moundville,
Alabama
X
X
X
X
King, Georgia
30
X
?
X
X
X
King, Georgia
223
X
?
X
X
X
Warren
Wilson, North
Carolina
16
X
X
X
X
X
X
Warren
Wilson, North
Carolina
29
X
X
X
X
X
Coweeta
Creek, North
Carolina
41
X
X
Coweeta
Creek, North
Carolina
43
X
X
Dallas Site,
East Tennessee
(7HA1)
45
X
X
Dallas Site,
East Tennessee
(7HA1)
67
X
X
Dallas Site,
East Tennessee
(8HA1)
136
X
X
DeArmond,
East Tennessee
26
X
X
X
DeArmond,
East Tennessee
49
X
X
X
X
X
X
14
Table 2 (continued). Archaeological Correlates Rattle Characteristics Checklist.
Burial
Number
Ceremonial
or Burial
Context
Eastern
Box
Turtle
Only
Carapace
or
Plastron
Modification
Residue
or
Staining
Rattle
Implements
Attached
to arm or
ankle
Hamilton
County,
East
Tennessee
n/a
X
X
X
X
Toqua,
East
Tennessee
150
X
X
X
Hiwassee
Island,
East
Tennessee
87
X
X
X
X
Hiwassee
Island ,
East
Tennessee
89
X
X
X
X
X
Hiwassee
Island,
East
Tennessee
118
X
X
X
X
X
Spiro,
Oklahoma?
X
Littlefield
site?
X
X
X
Mound
Bottom?
X
Garden
Creek,
North
Carolina
X
Turtle Rattle Types
There are three main constructions of rattles including body rattles7which were
commonly worn on the wearer’s ankle or upper arm handheld rattles, and wooden staff rattles.
Some rattles were made from copper in the shape of turtle shells at Mitchell which are turtle
substitute rattles (Sampson and Esarey 1993). The evidence of modification on the shells has
varied but distinct characteristics.
7 In all of the data compiled, body rattles are always on the ankle or upper arms; therefore, the term will be used as a
specific rattle type.
15
Body rattle. The first construction type is the body rattle, which is the most common rattle type
from the archaeological record in the Southeastern United States (see Table 2) and the
ethnographical/ethnohistorical data, discussed below. Typically, this type consists of a few
(approximately two) drilled holes in the carapace, as seen at Hiwassee Island (Lewis and
Kneberg 1970), to up to twenty holes, as evidenced in a 1920s example at the Oklahoma History
Center (Turtle 2009), then possibly tied with or hung by cordage (Voegelin 1942). The plastron
is also drilled at least once which is also evident at Hiwassee Island. However, un-modified and
non-drilled turtle shells can be used as rattles, merely by using cordage (Kevin Smith, personal
communication 2011). Un-modified and non-drilled turtle shells were tested through
experimental archaeology (Figure 4).
Figure 4. Turtle shell leg rattle without drilled holes.
Lewis and Kneberg (1970:126, Figure 28) provide an illustration of the Hiwassee rattles (see
Figure 2). The carapace has two drill holes; both are located between the third neural and third
costal on the right and left. The plastron only contains one drill hole, which is located in the
16
center of the hypoplastron (Lewis and Kneberg 1970:Figure 28). At the Archaic Period Ensworth
Site, the drill hole is located in the center on the hypoplastron (Deter-Wolf et al. 2004). Another
rattle carapace recovered from Hiwassee Island has four drill holes on the outer parts of the
costal creating a square and the marginals have been removed from the shell (Lewis and Kneberg
1970). The amount of drilled holes varies by group and site. Some shells show evidence of
polishing which could suggest that it had been worn or handled extensively (Brown 1967).
Several copper effigy turtle rattles were recovered from the Mississippian Period Mitchell Site in
Illinois (Sampson 1993). These are referred to as ―turtle substitute rattles.‖ These were used as
ankle/body rattles which are shown by the smaller drilled holes. The eastern box turtle is
typically used for this type of rattle.
Handheld rattle. The second construction type is the handheld rattle, which is common among
the Seneca and Iroquois. This type consists of several drilled holes along the edge in order to
fasten the carapace and plastron together. Next, a wooden handle is attached to the shell, which
runs parallel with the shell (Conklin and Sturtevant 1953:Figure 2; Parker 1909). The wooden
handles do not usually survive in the archaeological record. At the Vandal Cave burials, which
were located in Northeastern Arizona, one burial had a turtle shell rattle minus the handle. This
rattle was fastened together by a yucca fiber (Brown 1967). The animal’s neck skin is sometimes
stretched out in order to fit over the handle; and the turtle’s head is placed at the end of the
handle (Speck 1945). Snapping turtle is the most common species used for this type of rattle.
Wooden staff rattle. The third construction type is the wooden staff rattle, which is the most
common rattle type in the Western United States. This type is similar to the handheld rattle
except that it sits perpendicular on a long wooden staff (Figure 5). In some cases asphaltum was
17
used to better seal the gaps in between the carapace and plastron (Schneider and Everson 1989;
Schneider 1996; Strudwick and Koerper 2006). As previously mentioned, the use of asphaltum is
a Western residue, which is not readily available in the Southeastern United States. The desert
tortoise was typically used for this type of rattle.
Figure 5. California wooden staff rattle. Illustration by Joe Cramer with production assistance
from Kris Walden (LSA Associates, Inc.) in Strudwick and Koerper (2006:148, Fig. 4) after
Wallace 1980:106.
Southeastern United States Ethnographic and Ethnohistoric Evidence
Native American groups of the Southeastern United States, such as the Shawnee,
Seminole, Cherokee, and Creek still use turtle shell rattles or turtle substitute rattles. A turtle
substitute rattle, while not made from an actual turtle shell, echoes sounds similar to a real turtle
shell rattle and possesses a comparable meaning. Turtle shell rattles are predominantly associated
with adult females among the Seminoles, Cherokee, and Creek; however, shamans and leading
men use them in other tribes such as the Shawnee. The Shawnee use rattles that are connected
with ―magical songs,‖ typically sung by a shaman. The construction of the rattle for this use is
18
one of the handheld variety (Conklin and Sturtevant 1953:Figure 2). Often, quartz crystals are
placed inside in order to create the rattle effect (Voegelin 1942). For the Seminoles, turtle shell
rattles are occasionally seen8; however, in modern times, evaporated milk cans are employed
(Capron 1953) because of the decline in the turtle population (Dodd 2002).9
One of the earliest accounts of a rattle in the Southeast is noted by James Adair (1775),
who focused on the Catawbas, Cherokee, Chickasaw, and some Creek tribes, and gives a detailed
description of rattles. James Adair was born in Ireland and moved to South Carolina in 1735
(Hudson 1977:311). He was a trader, mostly of deer skin, with the Catawbas and Cherokee, and
later with the Chickasaws (Hudson 1977:311). Adair (1775:97) describes the rattles as being
filled with white pebbles and possibly beads. Adair (1775:180) also states that the rattles were
attached to a ―white deer-skin,‖ then were attached to both the outer part of the legs. Movarian
missionaries in 1803 witnessed a dance at Oostanaula, a Cherokee town. Relying on the
missionaries’ account, Schwarze (1923:79) notes, ―The last dance the missionaries witnessed
was done by women only, dancing around the pole, the men beating time. The female leader of
this dance wore leather shoes with turtle backs fastened thereto with which she mightily rattled!‖
The Cherokee use box turtle shells for the construction of turtle shell rattles (Fradkin 1990). This
group employed body rattles, which were fastened to the leg. Approximately five shells were put
on a hide (Figure 6), tied together, and the inside was filled with pebbles (Fradkin 1990). At this
time, there is no recorded information for the consumption of turtle by the Cherokee; however,
8 Many photos of Seminole rattles in the online archives of the American Museum of Natural History: Division of
Anthropology (http://anthro.amnh.org/north).
9 Kenneth Dodd (2002) discusses several instances where floods can cause major declines in the box turtle
population. Environmental factors could be at the source of the population decline.
19
they are known to use box turtles for medicinal purposes as well as medicine bags (Fradkin
1990:426).
Figure 6. Creek woman’s leg shackles or leggings which are attached to leather, Cleora and
Geneva Helbing Collection of Indian Arts and Crafts, Pope County Museum, Illinois,
http://popecountymuseum.files.wordpress.com/2011/05/1966-1-376y.jpg)
The box turtles were used to heal rheumatism and other ailments (Fradkin 1990:425).
Additionally, box turtles were used as a sacrifice by filling the shell with tobacco and throwing it
into a fire (Fradkin 1990). For some ritual purposes, the arms and legs of a box turtle were used
for the ―ball game,as it is believed the players would gain strength from the box turtle (Fradkin
1990). Zogry (2010:186) explains that participating in the anesto, or ball game, strengthens a
group’s (clan, tribe, etc.) identity. The ball is game is commonly associated with warfare,
especially in the preparation for battle. The term for ball game is also sometimes translated as
―little war‖ (Zogry 2010:186).
Creek women used leg turtle shell shackles, which is a form of body rattle. During the
Green Corn festival, the women engage in the ―Ribbon Dance‖ in which they will move in sync
20
to the sound of turtle shell rattles (Art Institute 2005). The Green Corn Festival comes about the
time that the corn turns ripe. It is a three-day festival, which no one eats the corn until it has been
blessed. Many activities are a part of the ceremony including: dance, ball game, busk, and ―black
drink‖ (Jessop 1997). The rattles are also used among the Stomp Dance (Conlon 2004).
A musical approach to the rattles incorporates sound and construction types/materials.
The reconstruction of sound quality and rhythm can be achieved through the reconstruction of
ancient rattles with the help of modern interpretations. The women strike a constant rhythm
while the men echo in song. There is the accentuation of a pounding noise as the river pebbles
bounce, twist, and turn violently within the shell. The noise reverberates so loudly that it
overtakes the voices of the singers (Conlon 2004:1). Some of the dances will last all night, until
sunrise (Conlon 2004).
Some Creek have a specific way of preparing the turtle after death. An anonymous
individual, who is part of the Creek Federation, told me, ―My grandmother would place the dead
turtle in an ant pile, which consisted of the large red ants, because the smaller ones would eat
through the connected pieces‖ (Anonymous, personal communication 2011).10 This process
allows for most of the interior and meaty portions to be removed, while keeping the connected
portions of the shell intact. Thus, the hinged portions on the plastron would stay connected, and
the plastron and carapace would also remain as one piece. This will allow the turtle to remain as
one entity without having to bind the pieces back together. On a side note, it would be very
10 I spoke with an anonymous individual who was part of the Creek Federation, while presenting a poster at the 2011
Southeastern Archaeological Conference.
21
difficult to prepare a turtle in this manner and use the interior portions of the turtle as food at the
same time.11
Archaeological Evidence from the Mississippian Southeast
Research on the late prehistoric archaeological records of Illinois, Arkansas, Alabama,
northwestern Georgia, western North Carolina, and Tennessee provide descriptions of turtle shell
rattles. This archaeological evidence derives from twelve sites including the Mitchell (Sampson
and Esarey 1993), Zebree, Moundville (Wilson and Davis 2003; Wilson et al. 2006), King,
Warren Wilson, Coweeta Creek, Dallas Site (7HA1), Dallas Site (8HA1), DeArmond, Hamilton
County, Tennessee Site, Toqua, and Hiwassee Island (see Table 1).
Mitchell, Illinois. The Mitchell site is located in Madison County, Illinois, and dates to
approximately 1150-1250 AD. Several copper effigy turtle rattles were found at this site12
(Sampson and Esarey 1993). The rattles fall into the classification of a substitute rattle, although
the copper may have originally covered an actual turtle shell. The placements of the perforations
make it a likely candidate for a rattle (Sampson and Esarey 1993). They were made from a thin
copper sheet that covered wood, bone, stone, or even possibly an actual turtle shell (Sampson
and Esarey 1993). From the Mississippian Period at Spiro, Oklahoma, a carved wooden turtle
rattle was overlaid with copper (Trevelyan 2004). These rattles may have been used for
ceremonial or ritual purposes; however, the purpose of the copper effigy rattles is unclear. The
rattles may have been used in dance or merely as an adornment (Sampson and Esarey 1993).
11 The limbs could be cut off before placing it in the ant bed, which could be used for food or broth for soup (Tanya
Peres, personal communication 2011).
12 A photo of a copper effigy rattle is available on the website of the Illinois State Museum, website listed in
references.
22
Zebree, Arkansas. The Zebree site is located in Mississippi County, Arkansas. A turtle carapace
rattle was found on a Middle Mississippian house floor (Figure 7) (University of Michigan 1970;
Morse and Morse 2009). Pebbles made a trail on the floor in between the carapace and plastron
(Dan Morse, personal communication 2011). For this rattle type, the dead turtle would likely
have been prepared in a similar way to the modern Creek, as discussed above.
Figure 7. Turtle shell rattle from Zebree (Permission granted by the University of Michigan
Museum of Anthropology, http://quod.lib.umich.edu/a/anthro1ic/x-27521/27521).
Moundville, Alabama. Moundville is located on the Black Warrior River in west-central
Alabama. The site is a very large Mississippian mound complex with ―twenty-six earthen
mounds‖ (Moundville 2011; Wilson et al. 2006). Turtle shell rattles were discovered in multiple-
household areas (Wilson et al. 2006). Clay pipes and other fineware were discovered in the same
context, which are commonly regarded as ritual items (Wilson et al. 2006). The context suggests
that the rattles were ritual items. Ceremonial life occurred in between the different households;
23
therefore, ritual and ceremony occurred outside of the large communal ritual areas and structures
(Wilson and Davis 2003; Wilson et al. 2006).
King, Northwestern Georgia. The King site is a late Mississippian town in northwestern Georgia.
There were two burials, Burials 30 and 223, which yielded turtle shells (Hally 2009). Burial 30
was identified as possibly female (Hally 2009:342). Two turtle shells of an unidentified species
were located on the right humerus and the other was adjacent to the left humerus. Several burial
items were placed in the grave, including: 12 points, a flintknapper kit, 4 bone tools, an antler
cylinder, a pulley-shaped ear spool, 2 turtle shells, and over 200 marine shell beads. The
flintknapper kit and antler cylinder indicate that a special skill was attributed to the individual.
Burial 30 also had the remains of a human infant, Burial 35, close to the feet of Burial 30 and it
was within the wall of the grave pit. This could be a mother and infant but there is no other
evidence of the mother and infant burial at the King site (Hally 2009:339). The absence around
the site of the mother and infant burial correlation around King suggests that an argument cannot
be made for the sex of Burial 30.
Burial 223 was a female, who was approximately 25 years old. Turtle fragments were
recovered from a plow scar that totally took off the left shoulder and upper arm (Figure 8).
Several burial items were placed in the grave including 23 points, 2 large bifacial blades,
hematite, one baculum (species not specified), one bracket type shell pin, one possible bone tool,
and one turtle shell. Seven of the artifacts are typically associated with male burials such as
points, bifacial blades, hematite, and baculum (all are associated with warfare, excluding the
baculum). Most bifacial blades are interred with males around 40 years old at King (Hally
24
2009:339). These seemingly male items being placed in a female grave will add to the discussion
of sex and gender below.
Figure 8. Burial 223 at King (Hally 2009: 9.1).
Warren Wilson, North Carolina. Warren Wilson is a late prehistoric stockaded village, located
on the former banks of the Swannanoa River on the Appalachian Summit in Southwestern North
Carolina. The occupation of the site is dated to approximately 1250-1500 AD (Rodning and
Moore 2010). There were two burials, 16 and 29, that yielded turtle shell rattle remains (Rodning
and Moore 2010:87). Burial 16 was a 42-year old (+/- 5 years) male, which places him in the
age group of elder according to Rodning and Moore (2010:87), who was buried with a number of
grave goods including: 20 shell beads, 25 conch shell fragments, 8 turtle shell rattle fragments,
24 rattle pebbles, and 25 quartz pebbles (Rodning and Moore 2010:87) (Figure 9). Dickens
26
Figure 11. Complete turtle carapace from Burial 29 at Warren Wilson (Photo courtesy of the
Research Laboratories of Archaeology, The University of North Carolina at Chapel Hill).
Coweeta Creek, North Carolina. Coweeta Creek is a late prehistoric and protohistoric Cherokee
town with a public structure and several domestic houses (Rodning and Moore 2010). This
village dates to approximately 1400-1700 AD (Rodning and Moore 2010:94). Burial 41 was a 23
year-old (+/- 3 years) female which yielded one turtle shell rattle and twenty-four shell bead
fragments. Burial 43 was a 17 year-old (+/- 3 years) adult female which had two turtle shell
rattles (Rodning and Moore 2010:94).
7HA1 and 8HA1, Eastern Tennessee. Lewis and Kneberg (1970:148) give the descriptions of
three burials that contained rattles. The two Dallas Phase sites are located in Eastern Tennessee.
Lewis (Lewis 1943:311) correlates the Dallas focus with protohistoric Muskhogean. At 7HA1
Dallas site, there were two burials, Burial 45 and 67, which yielded turtle shells. Burial 45 was a
12 to 15 year old of undetermined sex that had two turtle carapaces. Burial 67 was a 27 to 35
year-old female that had one turtle carapace. It was located around the rear of the skull and part
27
of the carapace was inside a pottery bowl. At the 8HA1 Dallas site, Burial 136 was a 16 to 18
year-old possible female who had a turtle carapace rattle of which the location was not specified
(Lewis and Lewis 1995:348) (see Figure 2).
DeArmond, East Tennessee. The DeArmond site is located in Eastern Tennessee and dates to
approximately 1200-1450 AD. Works Progress Administration archaeologists carried out
excavations in 1939-1940, followed by Wendell Walker in 1940 (McCarthy 2010). Burial 26
yielded an adult female with a turtle shell rattle. The female was fully flexed and facing a
northeastern direction (McCarthy 2010). Burial 49 also yielded an adult female that had a
perforated fragmentary turtle shell over her hands as well as beads13 under her hands (McCarthy
2010). Her body was partly flexed and facing a southwestern direction (McCarthy 2010). Both
burials were located in the same part of the village, Unit 2.
Hamilton County, Tennessee. The Hamilton County village site is located in Eastern Tennessee.
The grave of a 17 year-old female, who was part of the Dallas group, was excavated at the site.
The Dallas group lived in Eastern Tennessee around 1400-1500 AD (Lewis 1943). A turtle shell
rattle was interred with the female individual and was attached to her right arm (Lewis 1943)
(Figure 12).
13 The type of beads is not specified in the text.
28
Figure 12. Recreation of a Dallas woman, wearing turtle shell rattles on each arm (Lewis and
Kneberg 1970: Plate 103).
Toqua, East Tennessee. The Toqua site is located in Monroe County, eastern Tennessee. Burial
150 was a 16 to 20 year-old female was located in a tomb, dating to the late Mississippian Period
(ca. 1500 AD) (Chapman 1995). The burial was located on the summit of Mound A, and dating
to the latter phases of mound occupancy (Chapman 1995). There were several grave goods
including a short neck painted bottleof which the function is unknownmarine shell beads,
and turtle shell rattles on each ankle (Chapman 1995; Polhemus 1987).
Hiwassee Island, East Tennessee. Hiwassee Island is located in Meigs County, eastern
Tennessee. The site is a large island that sits at a crossroads of the left bank of the Tennessee
River and the Hiwassee. Turtle shell rattles are found in several burials at Hiwassee Island
29
(Lewis and Kneberg 1970). Burial 7, an adult female, had ten rattles located in a pile around the
lumbar region (Lewis and Kneberg 1970). Burial 89, a young female, had rattles positioned on
both upper arms (see Figure 12) (Lewis and Kneberg 1970). In Burial 118, an adult male had one
rattle placed on the right humerus (Lewis and Kneberg 1970). Two types of rattles were
discovered at the site. One type had two perforations through the carapace, which were located
on both sides between the third costal and third neural (see Figure 2). The perforation on the
plastron was located in the central part of the hypoplastron. Another version of a rattle has four
perforations around the outer parts of the carapace. On both sides, the perforations are located
around the second and sixth costals (Lewis and Kneberg 1970:126). For the first Hiwassee Island
rattle type, the dead turtle would likely have been prepared by placing it in an ant hill similarly
to how the anonymous individual’s grandmother prepared turtles, as discussed above. Only one
drill hole supports the fact that the plastron must have remained one entity.
Other possible sites. Holt (1996) re-analyzed the faunal remains of two sites, Sponemann and the
Dohack/Range site, which are located very close to Cahokia, Illinois. Several complete turtle
shells show signs of modification, which she concludes are possible rattles or bowls (Holt
1996:95, 100). At the Spiro site, Orr (1946: 235) associates caches of small pebbles with rattles.
At the Littlefield site, four burials had box turtle shells, which are interpreted as rattles (Orr
1946:243). At the Garden Creek site in southwestern North Carolina, one possible turtle shell
rattle was recovered from the burial of a male (Dickens 1976:129).
Discussion of the Archaeological Correlates
Location of turtle shell rattles. In the Mississippian Period, turtle shell rattles occur in eastern
Tennessee, southwestern North Carolina, and northwestern Georgia. Rattles have not been found
30
in burial contexts in Middle Tennessee despite extensive cemetery excavations by Gates P.
Thruston and through more modern resource management projects. Modern projects include
Gordon Town (Breitburg 1998), Rutherford-Kizer (Breitburg and Moore 2001), Brentwood
Library (Sichler and Moore 2005), and Averbusch (Romanski 1984). Prior to dam construction
on the Tennessee River, numerous sites, including burials, were excavated along the major rivers
in East Tennessee by the WPA archaeologists. In the 1800s, many burials were dug in Middle
Tennessee by individuals like Gates P. Thruston. Gates Phillips Thruston was a wealthy
Nashville businessman and Civil War general (Hale 1913; Young 1910) as well as a Fellow of
the American Association for the Advancement of Science; vice president of the Tennessee
Historical Society; a corresponding member of the New York and Philadelphia Archaeological
and Numismatic Societies (Young 1910:246).
In 1888, an Indian cemetery and town (A.D. 1050-1450) was discovered in the Nashville area on
the Noel Farm, owned by Dr. Oscar Noel (Smith 2010). Thruston directed the excavations of
approximately 3,000 to 4,000 burials on the farm (Hale 1913; Young 1910:248). Thruston (1890)
published a detailed analysis and comparison of various artifacts recovered from the burials.
These types of excavations mostly focused on the removal of complete artifacts although
carefully excavated. It is difficult to compare the quality of data and quantity of burials in Middle
Tennessee and East Tennessee (Aaron Deter-Wolf, personal communication 2011).
Another possibility is that the people of Middle Tennessee were different from the people
of eastern Tennessee. Rattles may be located in different contexts compared to the
burial/ceremonial context for east Tennessee turtle shell rattles. The context of turtle elements,
31
exclusively the carapace and plastron, should be investigated in order to determine the new
context(s).
Proposed rattle characteristics. Many of the sites exhibit several of the proposed characteristics
for turtle shell rattles (see Table 2). There are several biases of the literature that should be
discussed. Sometimes the exact species of turtle is not noted when discussing turtle shell rattles;
however, photographs can sometimes aid in confirming species identification. Carapace and
plastron fragments are noted but there is not a clear distinction if any other turtle elements,
excluding the carapace and plastron, were discovered in the same context. In the Southeastern
United States, residue and staining does not seem to occur (at least in this analysis). A re-analysis
of the specimens from the archaeological correlate sites could yield new results. Rattle
implements are also subjected to biases. River pebbles may have been discarded as soil during
excavations and seeds may not have preserved in the archaeological record.
Age and sex of burials. The sex of the archaeological correlate burials was predominantly female
(Table 3). Eleven of the sixteen (68.75%) burials are female and ten of the females fall into the
14 to 26 years old age category. The age of the archaeological correlate burials were
predominantly young (Table 4). Burial 67 at 7HA1 ranges from 27 to 35 years old. Two of the
sixteen (12.50%) burials are identified as possible female of which one is 16 to 18 years old.
Also, two of the sixteen (12.50%) burials are identified as male, which one is 42 +/- 5 years and
the other is an adult. The sex of the final burial is unknown (6.25%), but was assessed to be 12 to
15 years old at the time of death.
32
Table 3. Sex Distribution of Archaeological Burials.
Sex Total
Sex Total (%)
Female
11
68.75
Possible Female
2
12.50
Male
2
12.50
Unknown
1
6.25
Total
16
100.00
Table 4. Age Distribution of Archaeological Burials.14
12 to
26
12 to 26
(%)
27 to
35
27 to 35
(%)
37 to
47
37 to 47
(%)
Unknown
Unknown
(%)
Female
10
62.50
1
6.25
0
0.00
0
0.00
Possible Female
1
6.25
0
0.00
0
0.00
1
6.25
Male
0
0.00
0
0.00
1
6.25
1
6.25
Unknown
1
6.25
0
0.00
0
0.00
0
0.00
Total for Age
Group
12
75.00
1
6.25
1
6.25
2
12.50
Total Burials
16
New meanings and interpretations for turtles and turtle shell rattles could be pursued in
regards to sex and gender. According to the archaeological evidence, the burials with rattles are
predominantly young female. The possibility of a different gender could be presented, especially
given the account at the King site. The large quantity of grave goods, which some have ―male‖
characteristics (Hally 2009), could be evidence of a separate gender. Hally (2009:340) presents
information that could associate the females as ―War Woman‖ or ―Beloved Woman.‖ For the
Cherokee, the ―War Woman,who was revered, accompanied the men to war. At the stage of
post-reproduction, she reaches the title of ―Beloved Woman,‖ because war trips were no longer
possible; however, she was still held in high esteem (Hally 2009:341). Among modern Native
American tribes and communities, turtle shell rattles are predominantly worn by females. The
14 All percentages are out of the total number of burials.
33
evidence suggests that there is a close relationship between certain females and turtle shell
rattles. As the females embrace the rattles, semiotics surrounds the connection of person and
rattle; thus, the turtle shell rattle is the signifier and the signified has yet to be determined. The
deeper meaning of the signified could reveal new information about the relationship between
turtle shell rattles and their owners.
Experimental Archaeology and Turtle Shell Rattle Manufacture
Experimental archaeology is the process of recreating ancient artifacts or structures in
order to comprehend past technologies. The purpose of using experimental archaeology for this
project was to understand potential drilling methods and the effects on the turtle shell, breakage
patterns, the impact of river pebble on the shell, and tracing the modifications throughout the
construction process. By re-creating and documenting the manufacture process of one type of
turtle shell rattle, the body rattle, I was able to identify modifications on the modern specimens
that can be compared with those found in the archaeological record. The experimental
archaeology data was applied directly to the case study sites in relation to intentional
modification and rattles.
Methods
Approach and Design. From the ethnographic/ethnohistorical, archaeological, and modern
examples summarized above, I formulated a plan and design in order to create a turtle shell
rattle. The proposed characteristics were in part put to the test through extensive background
research and experimental archaeology. Specifically, four sites, Hiwassee Island (see Figure 2),
Zebree (see Figure 7), Apple Creek, and Ensworth, provided the main design for the
35
Figure 14. Experimental Plastron: (a) Hyoplastron hinge-tie; (b) Hypoplastron hinge-tie; (c)
Hyoplastron Central; (d) Hypoplastron Central.
The specimens.15 Dr. Peres obtained a Scientific Collector’s permit to obtain deceased box turtle
specimens, which was obtained through the Tennessee Wildlife Resources Agency. I searched
many different wooded areas around Nashville and Murfreesboro, fields in Castalian Springs,16
and on the side of the road. One was obtained on October 8, 2011, along US Highway 231 South
approximately ten miles north of Fayetteville, Tennessee. The specimen only had ants on it but
the carapace had been partially cracked. The specimen was frozen until it could be treated. the
turtle was treated by soaking it in a bucket of papain (papaya enzyme) and water, which was
changed every two days for about eight days. Papain breaks down meat tissues very quickly
(Ashie et al. 2002). Biz brand degreaser detergent, which was diluted with water, was used in
15 Only previously dead turtles were used for this project. No animals were harmed or killed for this project.
16 A couple of families allowed me to walk their pasture in search of turtles, which lie close to small streams and
wooded areas.
36
order to take off the remaining meat and tissue. The turtle remains sat in a mesh container for
approximately four days. Finally, the carapace and plastron were soaked in a diluted hydrogen
peroxide solution for about three hours in order to kill any remaining bacteria.
Another box turtle carapace was used in this experiment. This carapace was donated by a
student, Rae Smith, who had found the shell in the woods in Murfreesboro, Tennessee about two
years ago. Only the shell was recovered, as no other remains were located. Two other specimens,
one box turtle carapace and one other complete juvenile box turtle (although cracked), were
donated but have not been used at this point.
Documenting the process. The process was documented thoroughly through the use of a digital
camera and video camera. The digital camera is a Canon EOS Digital Rebel XTi. The video
camera is a Panasonic SDR-S7. The entire process of drilling and rattle creation was video
recorded in order to be reviewed. Digital photos were taken throughout the process, especially
after every few revolutions of the drill. Photos were also taken of breakage patterns and other
wear that occurred in the process. Measurements of the drilled holes were also taken at the end of
the drilling process. This process took place both in Nashville, Tennessee in a controlled
environment and in the Middle Tennessee State University Anthropology Lab. The drilling
process for each piece will be detailed below.
37
Turtle Shell Rattle Production
Drilling instruments. River cane, Arundinaria sp., and sand were first used to drill through the
shell (Figure 15).
Figure 15. Drilling Turtle Shell with River Cane.
River cane can be used to drill stone as well as other objects (Kevin Smith, personal
communication 2011). Larry Kinsella (1999) documented the process of drilling stone with a
river cane drill. The river cane was obtained from the banks of the Cumberland River. Different
sizes of both dry and green river cane were used in this first attempt. All failed to drill through
the turtle shell. A method of using a bow with the river cane was attempted which failed to
operate effectively. The last attempt with river cane was to begin a small hole with a chert drill,
then use the river cane. This was attempted because it made the river cane more stable; however,
this attempt also failed to drill the shell.
The second option was to use a bow and chert drill bit (Figure 16). This was the most
effective and efficient method. Drilling was attempted by hand as well as with the bow. The
bow was more efficient than by hand (Figure 17). For starting the drill cavity, it was more
accurate to place and start the hole by using my hands, instead of the bow.
38
Figure 16. Bow and chert drill.
Figure 17. Bow and chert drill in action.
Drilling the Carapace. Many things were brought to light through the drilling process. As I
began the drilling process, the walls started to form a concave shape. The interior wall of the drill
cavity has a very rough texture (Figure 18).
40
Figure 20. Circular shape from changing drill angle.
The river cane worked very well to polish out excess loose bone (Figure 21). The after-effects
are evident (Figure 22). As the shell was drilled from the other side, the opening expanded to
about the same size as the starting drill side. The opening retained a slight notch in the interior of
the drill opening (Figure 23).
Figure 21. Before using river cane to polish out excess bone.
41
Figure 22. After using river cane to polish out excess bone.
Figure 23. Ridge in the central part of the drill opening.
A total of ten holes were drilled in various parts of the eastern box turtle shell. First, a
marginal piece was drilled on the exterior (Figure 24). The drill area was approximately in the
top left corner. The piece exhibited the characteristics of the main description; however, it broke
42
as soon as the drill pierced the interior of the shell. The pressure towards the suture line and outer
edges were unable to endure the friction of the drill (Figure 25).
Figure 24. Experimental Marginal.
Figure 25. Experimental Marginal after break.
Second, Costal A also exhibited similar characteristics to the main drill description. The drill
angle was not changed throughout the process so the final product yielded an oval area (Figure
26).
43
Figure 26. Experimental Costal A.
Third, Costal B consisted of three costals that were still linked together (Figure 27). The costals
exhibited a few different qualities due to the location of drilling. The location was in between
two of the costals and close to the suture line. The first drill resulted in a break from a third
costal, which left a drill indentation on one side (Figure 28). The final drill on the costals resulted
in the sutures splitting apart at the point of perforation. There was a slight break in the side of the
drill opening. The largest measurement includes the slight break in the drill opening (Figure 29).
Figure 27. Experimental Costal B (Before any drilling).
44
Figure 28. Experimental Costal B after the split.
Figure 29. Experimental Costal B with break indentation.
Fourth, Carapace A resulted in a similar description (Figure 30). This one was also drilled from
the interior, which resulted in the suture coming apart. The original side creates uneven areas on
the interior side; therefore, the drill catches on the uneven areas on the interior, which breaks
apart the structure of the shell.
45
Figure 30. Experimental Carapace A.
Next, Carapace B and the plastron elements make up the complete rattle. Carapace B was
modeled after the Hiwassee Island archaeological evidence. The carapace was drilled in between
the third neural and third costal on both sides (Figure 31).
Figure 31. Experimental Carapace B.
46
Both drill openings maintained the structure of the main drill description. Neither of the openings
was drilled from the interior. For the right side, there were 125 turns by hand in order to start the
opening. The drill moves approximately 90 degrees by hand, whereas the bow turns nearly a
whole revolution. After the drill was turned approximately 106 times, the drill perforated through
the interior. A concave opening had formed after an additional 95 revolutions, although it was a
slight oval shape. An additional 295 turns were made by hand in order to create a more circular
opening. This process only took approximately five minutes, excluding the time spent on
documentation. For the hyoplastron and hypoplastron hinge-tie, and the hyoplastron central (see
Figure 14a-c), the openings were drilled from the interior but maintain the common
characteristics of the main drill description. Last, the hypoplastron was drilled in the central area
from the exterior (see Figure 14d), which was modeled after the three sites listed above (Figure
32). For the Ensworth rattle, the hypoplastron seemed to have a congruent opening instead of a
concave opening. It is possible that another instrument was used to polish out the interior of the
opening.
Figure 32. Experimental Hypoplastron Central.
47
Final measurements were taken for the ten drilled pieces (Table 5). For the original drill
side, all measurements were fairly consistent which ranged from 0.5 cm to 0.7 cm for the left-
right axis, excluding the hyoplastron (0.4 cm) central opening. Also, the anteroposterior axis
consisted of a range of 0.55 cm to 0.65 cm, excluding Costal B (0.45 cm) and hyoplastron central
(0.4 cm). For the other side, the left-right axis yielded a range of 0.35 cm to 0.5 cm, excluding
Costal A (0.6 cm). Also, the anteroposterior axis consisted of a range of 0.3 cm to 0.5 cm.
Measurements could vary depending on the size and type of drill.
Table 5. Final Measurements of the Ten Experimental Drill Openings.
Primary Drill Side
Secondary Drill Side
Left-
Right
(cm)
Anteroposterior
(cm)
Left-
Right
(cm)
Anteroposterior
(cm)
Drilled
from
Interior or
Exterior?
Marginal
0.6
Exterior
Costal A
0.7
0.6
0.6
0.4
Exterior
Costal B
0.5
0.45
0.4
0.2
Interior
Carapace A
0.7
0.6
0.5
0.4
Exterior
Carapace B
Left
0.7
0.65
0.4
0.35
Exterior
Carapace B
Right
0.7
0.65
0.45
0.35
Exterior
Hyoplastron
hinge-tie
0.5
0.55
0.4
0.45
Interior
Hypoplastron
hinge-tie
0.55
0.55
0.45
0.45
Interior
Hyoplastron
central
0.4
0.4
0.35
0.3
Interior
Hypoplastron
central
0.5
0.55
0.35
0.5
Exterior
Breakage patterns. An increased amount of drilling will also increase the probability of
breakage, as seen with Carapace A. The shell will begin to break or split apart close to the
sutures for the juvenile turtle carapace. The outcome could be different for a fully fused turtle
48
carapace. Therefore, this may not be the case for an adult turtle whose sutures have fully fused.
The outcome suggests that adult or mature turtles, which have completely fused sutures, are less
likely to break apart; thus, the adult specimens are more likely to be used as rattles. The pressure
from the drill can cause breakage. The results will sometimes leave very sharp angled fragments
(Figure 33). The pressure from the break can also result in indentations left by the drill (Figure
34). Sharp angles could be evidence for turtle shell rattles; however, other modified objects (such
as bowls, cups, medicine bags) or different construction methods may also cause similar
breaks.17 In conclusion, the chert drilling method for rattles can be identified by the combination
of sharp breaks and drill indentations.
Figure 33. Breakage patterns: sharp angle.
17 Sharp angles maybe produced from smashing a turtle, although no experiment has taken place, yet.
49
Figure 34. Drill indentation.
Rattle Implements. River pebbles (approximately 4-8mm), drum teeth, and hard seeds were items
used in order to create the rattle effect. I chose to use river pebbles for this study (Figure 35). The
pebbles were obtained from Stones River, located in Murfreesboro, Tennessee. After
approximately shaking the rattle 350 times, there were no immediate signs of unintentional
modification on interior of the carapace. Modern ethnographic studies reveal that the dances can
last overnight until the sunrise (Conlon 2004), which would supersede 350 shakes. Thus,
additional tests could reveal stress striations or other wear.
Figure 35. River pebbles collected from Stones River.
50
Cordage. The mostly likely type of cordage used was plant cordage, such as that made from
cedar. In this first experiment, suede cordage was used which was obtained from a local craft
supply store. There was no breakdown of the shell after 350 shakes, or noticeable signs of
breakage. There were also no immediate signs of wear.
Summary
Information on drilling techniques and practices as well as intentional and unintentional
modification was learned through this experimental archaeology project. The chert drill created a
concave, oval shape when drilled from one side of the shell. A notch is left in the interior of the
drill cavity when drilled from both sides. When the angle of the chert drill was changed 90
degrees, the oval shape changed to a concave circle. The pressure from the chert drill can cause
distinct breakage patterns, such as sharp angled breaks and may leave drill indentations. No
immediate unintentional modification was discovered in this experiment. The data gained from
experimental archaeology, along with the background research, informed the identifications and
interpretations of turtle remains at the two case study sites, Fewkes and Castalian Springs.
Case Study Sites: Fewkes and Castalian Springs
The background research and experimental archaeology was applied to the two case
study sites: Fewkes (40WM1) and Castalian Springs (40SU14), both located in Middle
Tennessee. The identified eastern box turtle remains from the two sites are used as
archaeological case studies to test the six indicators.
51
Fewkes (40WM1)
The Fewkes Site is located in Williamson County, near the Little Harpeth River (see
Figure 36). The area contains two distinct archaeological sites including the Mississippian Period
occupation and the historic Boiling Springs site. Joseph Jones gives the earliest reference to the
site in 1876 when depicting a clay figurine from the site (HCRS 1). William Edward Myer
conducted archaeological excavations in October 1920. Myer named the site after J.W. Fewkes,
Chief of the Bureau of American Ethnology who approved the excavations for Myer (Myer
1928). The Fewkes site contains five mounds in the midst of a village. In 1996, the Phase II
excavations produced many items including faunal remains and a possible connection to the
mound area (DuVall and Associates, Inc. 1997a). In 1998, the excavations entered Phase III.
Many of these remains have been analyzed by, or under the direct supervision of, Dr. Tanya M.
Peres. As of July 2010, the official total for the analyzed samples is 77, 892 vertebrate and
invertebrate specimens (ca. 56.94 kg)18; however, several thousand more have since been
analyzed as part of Dr. Peres’s zooarchaeology class at MTSU. Analysis of the faunal remains is
on-going.
18 The analysis is ongoing; therefore, the numbers are derived from the current excel spreadsheet totals.
52
Figure 36. Map of Fewkes and Castalian Springs (Photo courtesy of Dr. Tanya Peres).
Fewkes turtle re-analysis. The turtle remains included in this analysis were recovered during
Phase III excavations located west of the central mound complex (Peres 2010). A total of 167
turtles (94 are securely identified as eastern box turtle) are included in this analysis out of the
total 2,989 turtle remains19 and were recovered from several contexts (Features 55, 184, 549,
702, 818) (Table 6).20
19 After re-analyzing the approximately 3,000 turtle remains, the 167 turtle remains were chosen because they
exhibited modification or possible modification.
20 When possible, descriptions of features are given. Other feature numbers and test units are features of
indeterminate function or origin.
53
Table 6. Fewkes Associated Rattle Modifications (Features Only).
Feature 55
Feature 549
Feature 702
Feature 818
Emydidae
Terrapene
carolina
Testudines
Terrapene
carolina
Trionychidae
Partial Drill
(interior)
0
0
0
0
0
Partial Drill
(exterior)
0
1
2
0
0
Red Ochre
(interior)
1
0
0
0
0
Complete Drill
0
0
0
1
1
Worked
0
0
0
0
1
Snapped
0
0
1
0
0
Total
1
1
3
1
2
Feature 55 is located on the cusp of the palisade; it was identified as a possible borrow pit, which
was filled with ―domestic refuse‖ (Peres 2010:106). The area seems to be associated with food
refuse, possibly from Structure 21, due to the amount of low-quality deer bone and evidence of
marrow extraction (Peres 2010:106). In Feature 55, one marginal fragment identified as family
of water and box turtles (Emydidae) was noted as having red ochre appear on the interior (Peres
2010:107) (Figure 37). Also from Feature 55, one eastern box turtle marginal has been partially
drilled on the exterior (Figure 38).
Figure 37. Emydidae marginal stained with red ochre recovered from Feature 55 at Fewkes.
54
Figure 38. Partially drilled box turtle marginal recovered from Feature 55 at Fewkes.
It does not appear to have been drilled by a chert drill, although it is possible that taphonomic
agents acted on it post-deposition. Feature 184 is an upper fill associated with Burials 4 and 6,
and Feature 185 (which is a hearth area, positioned on top of an adult male burial, and a possible
feasting area) (Peres 2010:107-108). For Feature 184, one costal fragment shows evidence of
breakage. It exhibits a v-shape that is congruent with the experimental data (Figure 39). In
Feature 549, two turtle fragments, one costal and one plastron, have red ochre on the exterior
(Figure 40).
Figure 39. V-shaped break and cancellous portion of the shell recovered from Feature 184 at
Fewkes.
55
Figure 40. Stained with red ochre and exhibits sharp angles recovered from Feature 549 at
Fewkes.
The plastron fragment also exhibits a sharp angle, as if it had been broken during the production
process. In Feature 702, one eastern box turtle fragment has a semi-circle shape showing
evidence that is has been drilled. In comparison to the experimental data, it exhibits the sloping
inward pattern consistent with using a chert drill. In Feature 818, a soft-shell turtle
(Trionychidae) costal fragment has a slight puncture hole with a large hole that extends out. The
indentation exhibits the characteristics of using a chert drill, although the walls of the hole itself
are more oval shaped than a circular (Figure 41).
56
Figure 41. Drilled/Perforation in soft-shelled turtle recovered from Feature 818 at Fewkes.
Several other modified fragments fall into the category of other features of indeterminate
function or origin; therefore, these will be listed by their modifications (Table 7). Red ochre
appears thirteen times on different specimens, six of those are on eastern box turtle.
Table 7. Fewkes Associated Rattle Modifications (Including Features).
Emydidae
Terrapene carolina
Testudines
Trionyx ferox
Ground
0
6
0
0
Partial Drill
(interior)
0
0
0
0
Partial Drill
(exterior)
0
1
1
0
Polish
0
1
0
0
Indentation
(interior)
0
4
8
1
Indentation
(exterior)
0
3
4
0
Red Ochre
(interior)
1
2
4
0
Red Ochre
(exterior)
0
5
1
0
Worked
0
6
0
1
Complete Drill
0
1
1
1
Snapped
0
0
1
0
Total
1
29
20
3
57
A plastron fragment from Test Unit 42 exhibits breakage and has been worked on the interior
(Figure 42). This could also be evidence of grinding out the interior. For ground modifications, a
neural from Test Unit 43 has been ground out and the exterior exhibits polish.
Figure 42. Worked Turtle Shell recovered from Test Unit 42 at Fewkes.
Unit 100 has the most interesting evidence, the five elements make up the left and right
marginals 7 through the pygal, which all mend together as well as other carapace fragments. The
marginals have been ground and smoothed flat on the distal side (Figure 43). The grinding
pattern continues although it begins to taper off into the next marginals 7-8. These two sides still
exhibit a slight amount of grinding/polishing but are basically gone by the beginning of marginal
7. There is also a neural which mends to these parts, although it has not been ground smooth on
the inside. This type of modification is commonly associated with turtle shell bowls or cups;
however, neurals are typically ground, too. No wear patterns of this magnitude were discovered
during experimental archaeology.
58
Figure 43. Ground marginals recovered from Test Unit 100 at Fewkes.
For the partial drilled category, one eastern box turtle marginal has been drilled partially
on the exterior from Test Unit 38. This one exhibits very similar characteristics to the one in
Feature 55. For the complete drill category, two fragments from Test Unit 145 (Figure 44) and
Test Unit 11/13 are semi-circles that have been drilled. These exhibit very similar characteristics
to Feature 702. The majority of the break patterns on turtle shell, such as Feature 184 and Test
Unit 42, are on adult turtle shells.
Figure 44. Drilled turtle shell recovered from Test Unit 145 at Fewkes.
59
Castalian Springs (40SU14)
Castalian Springs is a Mississippian village complex dating to 1000 to 1400 AD. The site
is located close to a mineral spring and tributary of the Cumberland River (see Figure 36). The
village was composed of fortifications and several mounds the platform mound was 200 feet in
length and 11 feet in height (Smith and Miller 2009). It is one of four large villages around the
Nashville basin (Smith and Miller 2009). William Myer was the first person to excavate at the
site in 1891, 1893, and 1916-17 (Smith and Miller 2009). Dr. Kevin Smith led the first
professional excavations at the site from 2005 to 2011. Many faunal remains have been
recovered from the site (Smith and Miller 2009). The faunal remains have been analyzed by, or
under the direct supervision of, Dr. Tanya M. Peres. Analysis of the faunal remains is on-going.
Castalian Springs turtle re-analysis. A total of 32 turtles (5 of which are identified as eastern box
turtle) are included in this analysis out of the total 264 turtle remains and were recovered from
(Features 4, 9, 23) (Table 8).
Table 8. Castalian Springs Associated Rattle Modifications.
Terrapene carolina
Trachemys scripta
Ground
2
0
Polish
8
0
Complete Drill
0
1
Snapped
1
0
Total
11
1
Feature 4 was located in the area of a possible sweat lodge, which included the possible cleaning
out of a shrine (Kevin Smith, personal communication 2011). For Feature 4, a pond-slider
marginal fragment yielded a drilling indentation (Figure 45). It exhibits the sloping inward
motion as when using a chert bit drill. It had just broken through the exterior surface, when
60
drilling was halted. The other half of the circle is missing which could be evidence of breakage
during production. The break is old because older bone is typically darker and the pores of the
bone can hardly be clearly seen. In fresh breaks, the pores of the bone are very distinct. Next
from Feature 4, six eastern box turtle fragments exhibit polish of which five of them are polished
on the interior and exterior (Figure 46).
Figure 45. Drilled pond-slider turtle recovered from Feature 4 at Castalian Springs.
Figure 46. Polished turtle shell recovered from Feature 4 at Castalian Springs.
61
From Feature 9, the hypoplastron of an eastern box turtle exhibits signs that it has been ground.
On the interior middle, the porousness of the shell begins to show and it also retains a slight
polish. From Feature 23, an eastern box turtle costal fragment exhibits signs that it has snapped
during production. It is in the shape of a diamond with sharper ends and also exhibits a polish on
the exterior surface (Figure 47).
Figure 47. Breakage patterns: sharp angles recovered from Feature 23 at Castalian Springs.
Excavations during the 2011 field season yielded several complete box turtle carapaces.21
Two carapaces were located in Feature 119 (Figure 48). One was nearly complete; it measured
approximately 124.16 mm in length (anteroposterior), 87.51 mm in width (left-right axis), and a
height of 52.46 mm. The shell is slightly cracked and the interior is still filled with soil in order
to preserve any elements until lab analysis. There did not appear to be any modification from the
exterior but analysis could reveal otherwise. The second specimen is missing the outer edges of
the marginal and carapace. It measured approximately 83.28 mm in length (anteroposterior),
21 These specimens have yet to be fully analyzed in the lab but do merit some discussion. These will be tested for
any possible residues.
62
70.60 mm in width (left-right axis), a height of 32.45 mm, and the widest part (diagonally) is
79.60 mm. This one is filled with soil in the interior until further analysis. This specimen does
not have any visible modification from the exterior. These come from a borrow pit or trash area.
It seems to date to around the same time as Mound 3 construction (Kevin Smith, personal
communication 2011). There were some other turtle carapaces and fragments that come from
Mound 3. Some ritual objects, a partial Dover sword and part of a statue, and turtle carapaces
were scattered across the top of the mound before it was capped with soil (Kevin Smith, personal
communication 2011). They exhibit three of the characteristics for turtle shell rattles including
species, element, and ritual context; however, further analysis will be needed in order to have an
accurate interpretation of these shells. The majority of the break patterns, such as those recovered
from Feature 4, are on adult turtle shells.
Figure 48. Two complete box turtle carapaces from Feature 119 at Castalian Springs.
63
Discussion of Fewkes and Castalian Springs
The indicator and characteristic of burial context is lacking for both sites. At Castalian
Springs, many of the shells are in a ritual context. Myer does not note turtle shell remains in any
of the burials excavated at Castalian Springs (Kevin Smith, personal communication 2011).
There were no turtle remains noted by any excavators of Fewkes’ burials.22 Modifications on
turtle do occur but are fairly rare for both sites, and the modified turtle remains (listed above)
exhibit two to three of the six characteristics. Eastern box turtle was the majority of modified
remains; furthermore, carapace and plastron fragments were the dominant element. A few
modified pieces are a species other than box turtle, although the pieces do maintain the skeletal
representation and modification characteristics. From experimental archaeology, breakage
patterns reveal the possibility of rattle manufacturing because the breaks are on fully fused adult
turtle shells. However, the lack of drill indentations slightly decreases the probability of
identifying the broken turtle shell fragments as a rattle. Residue and staining does occur at
Fewkes but it does not at Castalian Springs. The residue is not a clear indicator as it has not been
noted in the archaeological correlates. Rattle implements do not show up at either site which
could be a sampling bias because the implements may have been discarded as soil or backdirt.
Thus, Fewkes and Castalian Springs show a lack of substantial evidence to support turtle shell
rattle production and use. Six possible interpretations are given for the lack of evidence (see
below); however, two possible biases, context and recovery in the excavation areas, will be
discussed first.
22 Aaron Deter-Wolf searched the NAGRA database for Fewkes burials; it mostly consisted of projectile points.
64
The interpretations are prefaced with a discussion of context and recovery in excavation
areas of the two sites. Context plays an important role in the discovery of turtle shell rattles. The
evidence suggests that the sites of Middle Tennessee and East Tennessee diverged in the
characteristic of context. Furthermore, the contextual divergence of Middle and East Tennessee
suggests that the areas are geoculturally distinct. This distinction is also made by Lewis
(1943:310) in regards to geography. Thus, for Middle Tennessee sites, turtle shell rattles may
have been deposited in other context, instead of burials. Rattles may have been manufactured
outside of the excavated areas. For the near absence of turtle shell rattles (as discussed above), I
offer six possible interpretations. First, the people of Fewkes and Castalian Springs were not
producing rattles on site. Second, they had refined their construction methods so that no mistakes
were made during production; thus, there were fewer failed turtle shell rattle prototypes. This
option, furthermore, could be evidence for the people holding turtles in high regard. If turtles
were considered sacred, then it may have been unacceptable to make mistakes in the production
of turtle shell rattles. The Cherokee, for example, used turtle for ritualistic purposes (Fradkin
1990). Third, turtle rattles only occur in particular burial or ceremonial contexts. Fourth, it is
possible that the rattles were associated with some kind of status that no one achieved at these
two sites, or that the rattles represented a particular meaning that was not significant to the
people. Fifth, the people were not constructing new rattles but were simply passing them down
through families or through another type of non-bloodline lineage such as to the holder of a
specific status. Last, the people were utilizing unmodified turtles and turtle rattles.
Many of the fragments listed above fall into the intentional and unintentional modified
categories. Future investigations should look at other possible uses of turtle, excluding turtle
65
shell rattles, in order to explain the presence of turtle fragments. As briefly discussed above,
turtle shell bowls, cups, spoons, medicine bundles/bags, sacrifices, and subsistence have all been
discovered in the archaeological record or through ethnography. If turtles were not used for
subsistence, then the possibility arises that the people were in close communion with the turtles,
whether they were alive or dead. However, the evidence does suggest that the turtles did acquire
some level of meaning and purpose by entering the cultural realm.
Discussion and Summary
Discussion and Future Research. Several new questions have formed through this research,
which expands far beyond the scope of this paper. Subsistence is a very interesting topic in
regards to turtle. Many have cited Carr (1952), noting that people became sick after eating box
turtle. Turtles eat poisonous mushrooms at not harm to themselves, but this causes the turtle flesh
to have a level of toxicity that humans cannot tolerate. There have been no known scientific tests
on this matter or even the possible differences between Euro-Americans and Native Americans.
This concept must be tested. Another way to test it in the archaeological record is to look at the
non-shell turtle elements, which exclude the carapace and plastron, and see if they appear in food
refuse areas. On a different food-related note, a belief in the sacred is not necessarily always
upheld; therefore, the people of Fewkes and Castalian Springs could possibly have been violating
the sacredness of the turtle by eating them.
More experimental archaeology needs to be done in order to gain a larger sample
and a broader understanding of the process of rattle manufacture and the appearance of
manufacturing by-products. This can happen in a variety of different ways, such as using
66
different drilling instruments and construction models. One addition could be placing a few
rattles on a hide then attaching to the leg or using a handle rattle type. Second, the type of
cordage, mainly cedar cordage, should be pursued in order to determine any unintentional
modification. Third, tests should be done using pine pitch or resin or other available glue-
like materials in order to better understand the process and lasting effects on the turtle shell.
Glue-like materials do not show up in the archaeological evidence; however, the materials
may not have been detected during the analysis of the turtle fragments. Fourth, new ways to
prepare a deceased turtle should be tested, such as placing the turtle in an ant pile or
scraping out the inside of a turtle with an antler tool or other tools. Then, the modification
effects on the shell would be apparent. Additional tests and new strategies using river cane
should be attempted in order to be sure that it is in fact not a viable method. Also,
additional tests should be sought after for breakage patterns. During the drilling process,
breaks can cause sharp angles but simply smashing a shell may have the same effect.
Additionally the rattle implements, such as river pebbles, drum teeth, and hard seeds,
should be further tested in order to determine the effects on the shell. Another question to
pursue is what it looks like when a break occurs around a modified section like a drill
opening or cavity The modified areas of the shell should be purposefully broken in order to
determine breakage patterns. This could be done by putting too much pressure with the
drill. Further experimental data could yield new results and possibilities.
A more musical approach could also be taken in order to better understand the aesthetics
of different cultures. This approach would allow for the reconstruction of sounds of the ancient
peoples. Sound quality could be tested for the various construction types. A few different
67
research questions could be pursued. Does the number of holes make an acoustical difference?
Does different cordage or placing them on hide diminish the sound quality? Sound qualities
could be tested by focusing on the acoustics of different elements (drum teeth, pebbles, and hard
seeds), amount of drilled holes, construction types, species, and placement. The sound of copper
rattles versus bone rattles could also be compared. A more Ethnoarchaeological approach could
be taken in order to better understand how rattles are used, constructed, and perceived today in
order to better understand the past.
Other areas of modified and unmodified turtle use should be pursued as well. Turtle shell
bowls, cups, scoops, or spoons would be an interesting experimental archaeology project as well.
Modifications that are usually associated with those artifacts are scoring and snapping, ground
marginals, polish, ground neurals, etc. Next, many traditional natural transformations need to be
tested in order to be certain they are not cultural transformations. It would be interesting to have
a normal box turtle shell interred in the ground with one that has been used as a rattle. New
evidence could emerge for how differently the shells break down in the soil. Does the rattle
become more porous? Do modifications become more/less distinct? There are many small
indentations on some shells, which may simply be naturally occurring transformations; however,
the indentations could be from an intentional modification. Modern specimens should be
compared to the archaeological specimens in order to ascertain these indentations.
There are also a substantial amount of burned turtle fragments, which raises several
questions. Were the shells used for sacrificial purposes, such as the Cherokee have been known
to do (Fradkin 1990)? Were the remains parts of food refuse? Can the shell exhibit polish after
burning? Would they burn the shell in order to create color? Would they cook inside of it or use
68
as another food related container? Further research and tests should be run on the use of red
ochre. Can red ochre turn the shell a different color? Was red ochre used for color or part of a
binding agent? Were turtle shells, red ochre, and tattooing associated in any way? Can any type
of coloring or binding agent cause a break down in the shell?
Stable isotope analysis could be used in order to determine whether the turtles were local
or non-local in origin. Were the people catching the turtles locally or getting them from
somewhere else? Have the turtles or rattles been passed down through families (if they have
migrated from the original position, then this would show up in the analysis)? Are the turtles
living in close proximity or even within the village? Future research and analysis could yield
new evidence for unmodified and modified uses of turtle.
Summary. Turtles had a wide range of uses throughout prehistory. The categories of modified
and unmodified remains were introduced, as well as their sub-categories, which are intentional
and unintentional modification (modified), and non-subsistence and subsistence (unmodified).
The focus of this study was on the intentional modification of turtle shells in the production and
manufacture of rattles, as well as some unintentional modification in regards to the experimental
rattle. Through extensive background research and experimental archaeology, six characteristics
are proposed for identifying turtle shell rattles in the archaeological record, which are context,
turtle species, turtle skeletal representation, modifications, residue/staining, and rattle
implements. Residue/staining is the least likely to be a rattle characteristic in the Southeast
because it is not noted in any of the archaeological correlate specimen data; however,
experimental archaeology and possible re-analysis of the archaeological correlate specimens
could yield new results. The other five are clearly seen in the archaeological evidence and the
69
ethnographic/ethnohistoric data. During the Historic and Late Prehistoric periods, turtle shell
rattles are found mostly in ceremonial and burial contexts. Eastern box turtle is the majority of
the identified turtle species used in rattle manufacturing. Carapace and plastron make up most of
the turtle skeletal representation.
The experimental activity allowed me to discover the processes that occur in between the
acquisition of the raw material and the final product. I created turtle shell rattles that used similar
materials and construction types to the background data. Holes drilled by a chert drill can be
identified by the concave circle/oval. The Lewis and Kneberg (1946:Figure 28) illustration of the
Hiwassee Island rattle proved to be an accurate position for drilling a carapace and plastron. By
studying the drilling stages, we can identify these modifications among artifacts in the
archaeological record. The breakage patterns among turtle shell allow us to see and identify this
process. The study of cordage and river pebble can also allow us to see the unintentional
modification upon turtle shell.
Fewkes and Castalian Springs both yielded two to three characteristics per specimen.
Both sites seem to be lacking the most common factors such as burials. The rattles maybe
located in different contexts for Middle Tennessee. It is possible that the modified turtles were
part of the production process. A few drilled elements correspond with the chert drill in the
experimental data. Fewkes yielded a few of the characteristics that are congruent with the
indicators of turtle shell rattle production such as breakage patterns, polish, staining of red ochre,
ground, and drilled; however, the ceremonial/burial context is lacking for the modified remains.
70
New case sites in Middle Tennessee should be chosen in order to better test the proposed turtle
shell rattle characteristics and look for new rattle contexts.23
23 I have not looked at the specific data of other Middle Tennessee sites, aside from Fewkes and Castalian Springs,
but will pursue in future research. Archaic sites include Ensworth and Fernvale. The indicators for two
Mississippian sites, Brentwood Library project and Rutherford-Kizer, could be tested and compared to the case
study sites presented in this paper.
71
References Cited
Adair, James
1775 The history of the American Indians. Edward & Charles Dilly, London. Electronic
Document,
http://www.archive.org/stream/historyofamerica00adairich#page/n3/mode/2up, accessed
October 20, 2011.
American Museum of Natural History: Division of Anthropology
1910 Knee Rattle: North American Ethnographic Collection. American Museum of
Natural History: Division of Anthropology. Electronic document,
http://anthro.amnh.org/anthropology/databases/common/image dup.cfm?catno=50%2E1
%2F%202328&curr_page=&from_anthro=yes, accessed October 12, 2011.
Art Institute of Chicago
2005 American Indian Ancient Art in the Midwest and South. Electric document,
http://www.artic.edu/aic/collections/citi/resources/Rsrc_001118.pdf, accessed September
5, 2011. Art Institute of Chicago.
Ashie, I., Sorensen, T., and Nielsen, P.
2002 Effects of Papain and a Microbial Enzyme on Meat Proteins and Beef Tenderness.
Journal of Food Science, 67: 21382142.
Babcock, H.L.
1919 Turtles of New England. Memoirs of the Boston Society of Natural History,
8(3):325-431.
Boyd, C. Clifford, Jr., Donna C. Boyd, Michael B. Barber, and Paul S. Gardner
2005 The Bonham Site (44SM7): A Late Woodland Village Complex in Smyth County,
Virginia. Virginia Department of Historic Resources Research Report Series No. 16,
Richmond.
Breitburg, Emanuel
1998 Chapter IX. Faunal Remains. In Gordontown: Salvage Archaeology at a
Mississippian Town in Davidson County, Tennessee, edited by M. C. Moore and E.
Breitburg, pp. 147-168. Research Series No. 11, Division of Archaeology, Tennessee
Department of Environment and Conservation, Nashville.
Breitburg, E., and M. C. Moore
2001 VIII. Faunal Remains. In Archaeological Excavations at the Rutherford-Kizer Site:
A Mississippian Mound Center in Sumner County,Tennessee, edited by M. C. Moore and
K. E. Smith, pp. 119-133. Research Series No. 13, Division of Archaeology, Tennessee
Department of Environment and Conservation, Nashville.
72
Brown, Donald Nelson
1967 The Distribution of Sound Instruments in the Prehistoric Southwestern
United States. Ethnomusicology 11:71-90.
Capron, Louis
1953 The medicine bundles of the Florida Seminole and the Green Corn Dance. US
Government Printing Office, Washington.
Carpenter, Edmund
1942 Iroquoian Figurines. American Antiquity 8:105-113.
Carr, Archie
1952 Handbook of turtles; the turtles of the United States, Canada, and Baja
California. Comstack Publishing Associates, New York.
Chapman, Jefferson
1995 Tellico Archaeology: 12,000 Years of Native American History. Revised Edition.
The University of Tennessee Press, Knoxville.
Conklin, Harold C. and William C. Sturtevant
1953 Seneca Indian Singing Tools at Coldspring Longhouse: Musical Instruments of the
Modern Iroquois. Proceedings of the American Philosophical Society 97(3): 262-290.
Conlon, Paula
2006 Alligator Clans in Oklahoma: Stomp dancing in Indian Territory. Indigenous Policy
Journal 17(2): 1. Electronic document,
http://indigenouspolicy.org/Articles/VolXVIINo2/AlligatorClansinOklahomaCreekSemin
oleStomp/tabid/185/Default.aspx, accessed on November 20, 2011.
Cook, Robert Allen
2007 Sunwatch: Fort Ancient Development in the Mississippian World. University of
Alabama Press, Tuscaloosa.
Cridlebaugh, Patricia A.
1986 Penitentiary Branch: A late Archaic Cumberland River Shell Midden in Middle
Tennessee, No. 4. Tennessee Department of Conservation, Divison of Archaeology,
Nashville, Tennessee.
Deter-Wolf, Aaron, Sean Norris, Marc Wampler, Josh Tuschl
2004 The Ensworth School Project: Archaeological Investigations at Site 40DV184,
Davidson County, Tennessee. TRC, Inc., Nashville, Tennessee.
Dickens, Roy S.
73
1976 Cherokee Prehistory: The Pisgah Phase in the Appalachian Summit Region.
University of Tennessee Press, Knoxville.
Dodd, C. Kenneth
2002 North American Box Turtles: A Natural History. University of Oklahoma Press,
Norman, Oklahoma.
DuVall and Associates, Inc.
1997a Preliminary Report: Phase II Archaeological Testing of the Fewkes Site (40WM1),
State Route 441 (Moore's Lane), Williamson County, Tennessee. Report submitted to
Tennessee Department of Transportation, Nashville.
Fontaine, Roy
2011 Making Pitch Glue. Electronic Document,
http://www.survivaltopics.com/survival/making-pitch-glue, accessed November 10,
2011.
Fortier, Andrew C, Thomas Maher, and Joyce A Williams
1992 The Sponemann Site: The Formative Emergent Mississippian Sponemann Phase
Occupations (11-Ms-517). American Bottom Archaeology, Volume 23.
Fox, Andrew, Carl Heron, and Mark Q. Sutton
1995 Characterization of Natural Products on Native American Archaeological and
Ethnographic Materials from the Great Basin, USA: A Preliminary Study. Archaeometry
37(2):363-375.
Fradkin, Arlene.
1990 Cherokee Folk Zoology: The Animal world Native American People (1700-1838).
Garland Publishing, Connecticut.
Gibby, Evard H.
1999 Making Pitch Sticks. In Primitive technology: a book of earth skills, edited by
David Wescott, pp. 189-190. Gibbs Smith, Utah.
Griffin, James B.
1967 Eastern North American Archaeology: A Summary. Science 156: 175191.
Hale, Will T.
1913 A History of Tennessee and Tennesseans: The Leaders and Representative Men in
Commerce, Industry and Modern activities. Lewis Publishing Company, Chicago.
Hally, David J.
2009 King: The Social Archaeology of a Late Mississippian Town in Northwestern
Georgia. University of Alabama Press, Tuscaloosa.
74
Heritage Conservation and Recreation Service (HCRS)
1971 National Register of Historic Places Inventory Nomination Form: Description.
Item 7: 1-2.
Holt, Julie. Z.
1996 Beyond Optimization: Alternative Ways of Examining Animal Exploitation. World
Archaeology 28:89-109.
Hudson, Charles
1977 James Adair as Anthropologist. Ethnohistory 24(4): 311-328
Illinois State Museum
1991 Copper tortoise shell effigy rattle from the Mitchell Site. Electronic document,
http://www.museum.state.il.us/RiverWeb/landings/Ambot/prehistory/archives/images/art/
pages/rattle.html, accessed May 2011.
Ingersoll, Daniel W., John E. Yellen, and William Macdonald.
1977 Experimental archaeology. New York: Columbia University Press.
Jessop, Erin
1997 Green Corn Festival. Electronic Document,
http://teacherlink.ed.usu.edu/TLresources/units/Byrnes-celebrations/corn.html,
accessed on November 21, 2011.
Kinsella, Larry
1999 Drilling Stone. In Primitive technology: a book of earth skills, edited by David
Wescott, pp.92-93. Gibbs Smith, Utah.
Keel, Bennie, C.
1976 Cherokee Archaeology: A study of the Appalachian Summit. The University of
Tennessee Press, Knoxville.
Kelly, John E.
2006 The Ritualization of Cahokia: The Structure and Organization of Early Cahokia
Crafts. In Leadership and Polity in Mississippian Society, edited by Brian M. Butler and
Paul D. Welch, pp. 236-263. Center for Archaeological Investigations Occasional Paper
No. 33. Southern Illinois University, Carbondale.
Langenwalter II, Paul E.
1983 Bone Artifacts. In Archaeological Studies at Oro Grande, Mojave Desert,
California, edited by C.H. Rector, James D. Swenson, Philip J. Wilke. San Bernardino
County Museum Association.
75
Lewis, Thomas M.
1943 Late Horizons in the Southeast. Proceedings of the American Philosophical Society
Vol. 86(2): 304-312.
Lewis, Thomas and Mandeline Kneberg
1970 Hiwassee Island: An Archaeological Account of Four Tennessee Indian Peoples.
University of Tennessee Press, Knoxville.
Lewis, Thomas and Mandeline Lewis
1995 Prehistory of the Chickamauga Basin. University of Tennessee Press, Knoxville.
Markuson, Kevin
1998 The People of Mound Bottom: The Mississippian Culture. The Advocate 8:6.
McCarthy, Donna M.
2011 Using Osteological Evidence to Assess Biological Affinity: A Re-evaluation of
Selected Sites in East Tennessee. PhD dissertation, Electronic Document,
http://trace.tennessee.edu/cgi/viewcontent.cgi?article=1977&context=utk_graddiss,
accessed September 1, 2011. University of Tennessee.
Morse, Phyllis and Dan F. Morse
1990 The Zebree Site: An Emerged Early Mississippian Expression in Northeast
Arkansas. In The Mississippian Emergence, edited by Bruce D. Smith, pp. 5166.
Smithsonian Institution Press.
Morse, Daniel F. and Phyllis A. Morse
2009 Archaeology of the Central Mississippi Valley. University Alabama Press,
Tuscaloosa.
Moundville Archaeological Park
2011 Ancient Site. Electronic document, http://moundville.ua.edu/?page_id=16, accessed
December 17, 2011.
Myer, William E.
1928 Two Prehistoric Villages in Middle Tennessee. Forty-first Annual Report of the
Bureau of American Ethnology,1919-1924 485-614. Washington.
Orr, Kenneth G.
1946 The Archaeological Situation at Spiro, Oklahoma: A Preliminary Report. American
Antiquity 11:228-56.
Parker, Arthur C.
76
1909 Secret Medicine Societies of the Seneca. American Anthropologist 11: 161-
85.
Parmalee, Paul W., Andreas A. Paloumpis, and Nancy Wilson
1972 Animals Utilized by Woodland Peoples Occupying the Apple Creek Site, Illinois.
Illinois State Museum, Springfield Illinois.
Pechanga
2010 Rattlemaking. Electronic document, www.pechanga-nsn.gov/page?pageId=17,
accessed on February 9, 2011.
Penders, Thomas
2002 Bone, Antler, Dentary, and Lithic Artifacts. In Windover: Multidisciplinary
Investigations of an Early Archaic Florida Cemetery. University Press of Florida,
Gainesville, FL.
Peres, Tanya M.
2010 Zooarchaeological Remains From The 1998 Fewkes Site Excavations,
Williamson County, Tennessee. Tennessee Archaeology 5(1):100-125.
Polhemus, Richard
1987 The Toqua Site 40MR6, Vol. 1-2. The Tennessee Valley Authority, Knoxville,
Tennessee.
Pope County Museum
2011 Creek Woman’s Leg Shackles or Leggings. In Cleora and Geneva Helbing
Collection of Indian Arts and Crafts. Electronic Document,
http://popecountymuseum.files.wordpress.com/2011/05/1966-1-376y.jpg, accessed
October 2011.
Reilly, F. Kent and James F. Garber
2007 Ancient Objects and Sacred Realms: Interpretations of Mississippian Iconography.
University of Texas Press, Austin, Texas.
Research Laboratories of Archaeology
1990 Warren Wilson NAGPRA Inventory. Courtesy of the Research Laboratories of
Archaeology, The University of North Carolina at Chapel Hill.
Research Laboratories of Archaeology
2005 Specimen Catalog (accession nos. 2020 to 2093). Electronic document,
http://rla.unc.edu/Collections/RLA Specimen Catalog (2020-2093).pdf, accessed on
October 26, 2011.
Rodning, Christopher B., and David G. Moore
77
2010 Mortuary Practices in Late Prehistoric and Protohistoric Southwestern North
Carolina. Southeastern Archaeology 29:80-100.
Romanoski, Stephanie M.
1984 Analysis of the Faunal Remains from the Averbuch Site. In Averbuch: A Late
Mississippian Manifestation in the Nashville Basin, Volume I, edited by Walter E.
Klippel and William M. Bass, pp. 13.1-13.27. Copy on file, Tennessee Department of
Environment and Conservation, Division of Archaeology, Nashville.
Sampson, Kelvin and Duane Esarey
1993 A Survey of Elaborate Mississippian Copper Artifacts from Illinois. Illinois
Archaeology 5:452-480.
Schneider, Joan S.
1996 The Desert Tortoise and Early Peoples of the Western Desert.
http://www.tortoise-tracks.org/publications/schneider.html. Accessed February 15,
2011.
Schneider, Joan S. and G. Dicken Everson
1989 The Desert Tortoise (Xerobates agassizii) in the Prehistory of the
Southwestern Great Basin and Adjacent Areas. Journal of California and Great
Basin Anthropology 11:175-202.
Schwarze, Edmund
1923 History of the Moravian Missions among Southern Indian Tribes of the
United States. Moravian Historical Society, Transactions, Special Series 1. Time
Publication Company, Bethlehem, Pennsylvania.
Sichler, Judith and Michael C. Moore
2005 IX. Faunal Remains. In The Brentwood Library Site: A Mississippian Town on
the Little Harpeth River, Williamson County, Tennessee, edited by Michael C.
Moore, pp. 199-208. Research Series No. 15. Tennessee Department of
Environment and Conservation, Di-vision of Archaeology, Nashville.
Smith, Kevin E.
2010 Gates P. Thruston Collection of Vanderbilt University. Electronic document,
http://tennesseeencyclopedia.net/entry.php?rec=1382, accessed December 19, 2011.
Smith, Kevin E., and James V. Miller
2009 Speaking with the ancestors: Mississippian Stone Statuary of the Tennessee-
Cumberland Region, 67-72. University of Alabama Press, Tuscaloosa.
78
Society for Georgia Archaeology (SGA)
2007 Points, Pottery, Hafting. The Society for Georgia Archaeology.
http://thesga.org/tag/aboriginal-lithics/. Accessed March 23, 2011.
Speck, Frank G.
1945 The Iroquois : A Study in Cultural Evolution. Cranbrook Institute of Science,
Bloomfield Hills, Michigan.
Strudwick, Ivan H. and Henry C. Koerper
2006 A Unique Turtle Effigy from CA-ORA-269. Proceedings of the Society for
California Archaeology 19:145-150
Texas Archaeological Research Laboratory
2008 Caplen Mound. Electronic document,
http://www.texasbeyondhistory.net/coast/images/ap4.html, accessed February 15, 2011.
2010 Horn Shelter: A Paleo-Indian Grave. Electronic document,
http://www.texasbeyondhistory.net/horn/burials.html, accessed November 10, 2011.
Thruston, Gates P.
1890 The Antiquities of Tennessee and Adjacent States. Robert Clarke and Company,
Cincinnati.
Townsend, Richard F.
2004 Hero, Hawk, and Open Hand: American Indian Art of the Ancient Midwest and
South. Yale University Press, Connecticut.
Trevelyan, Amelia M.
2003 Miskwabik, Metal of Ritual: Metallurgy in Precontact Eastern North America.
University Press of Kentucky, Lexington, Kentucky.
University of Michigan Museum of Anthropology
1970 Turtle Carapace Rattle. Electronic document,
http://quod.lib.umich.edu/a/anthro1ic/x-27521/27521, accessed September 20, 2011.
University of Michigan Museum of Anthropology, Ann Arbor.
Voegelin, E. W.
1942 Shawnee Musical Instruments. American Anthropologist, 44: 463475.
Wallace, William J.
1980 A Turtle Shell Rattle from Long Beach. The Masterkey 54:102-
107.
Walthall, John
79
1990 Prehistoric Indians of the Southeast: Archeology of Alabama and the Middle South.
The University of Alabama, Tuscaloosa.
White Deer, G.
1995 Pretty Shellshaker. In Remaining Ourselves: Music and Tribal Memory. Oklahoma
City, Oklahoma: State Arts Council of Oklahoma, 10-12
Wilson, Gregory D. and R. P. Stephen Davis, Jr.
2003 Mapping the Moundville Community: A GIS Analysis of the Moundville Roadway.
Research Laboratories of Archaeology, UNC-Chapel Hill. Electronic document,
http://rla.unc.edu/Mdvlfiles/paper/Wilson-Davis%202003%20SEAC.pdf, accessed on
September 1, 2011.
Wilson, Gregory D., Jon Bernard Marcoux, and Brad Koldehoff.
2006 Square Pegs in Round Holes: Organizational Diversity between Early Moundville
and Cahokia. In Leadership and Polity in Mississippian Society, edited by B. M. Butler
and P. D. Welch, pp. 4372. Occasional Paper No. 33. Center for Archaeological
Investigations, Southern Illinois University, Carbondale. Bostrom, Peter and Illinois State
Museum.
Young, B.E.(editor)
1910 General Gates P. Thruston. Vanderbilt University Quarterly 10(4):245-274.
Zogry, Michael J.
2010 Anetso, the Cherokee Ball Game: At the Center of Ceremony and Identity.
University of North Carolina Press.
... The archaeological record indicates that, across eastern North America, turtle shell rattles were associated with ancient rituals and ceremonies, as evidenced by the archaeological contexts of rattles (Pearce 2005; see also Brown 2011;Peres 2017, 2018; Supplementary Table 1). Additionally, turtle shell rattles recovered from burial contexts were presumably buried with the person that had used the rattles (e.g., Swanton 1928a:396). ...
... Turtle shell rattles were used throughout the Woodland period (ca. 1000 BC -AD 1000) (Brown 2011). For southeastern Mississippian period turtle shell rattles, about 90% of the turtle shell rattles were recovered from burial contexts (Brown 2011). ...
... 1000 BC -AD 1000) (Brown 2011). For southeastern Mississippian period turtle shell rattles, about 90% of the turtle shell rattles were recovered from burial contexts (Brown 2011). ...
Article
Full-text available
Throughout North America, from the Archaic period (ca. 8000–1000 BC) to the present, Indigenous Peoples used turtle shell rattles in a variety of cultural contexts, including in ceremonies. As a material, turtle shells can be an abundant, accessible, and easily processed raw material, whose shape and size lends itself to be a musical instrument. Many Indigenous Peoples in North America have cosmological, foundational beliefs about turtles. These beliefs provide a greater understanding of why turtle shell rattles are incorporated into ceremonies and dances. Furthermore, they help explain why they are used to keep rhythm, which in turn provides a basis for spiritual energy and experience. This paper reviews the ethnographic, ethnohistoric, and archaeological records of turtle shell rattle music in the contiguous United States. Turtle shell rattles are related to spiritual concepts of sound and are culturally defined and contextualized. A comprehensive review of their use reveals insights into the musical knowledge of several ancient and historic communities.
... Drums and rattles fall into this category of instruments, and the latter are the focus of this paper. Across North America, rattles fashioned from turtle shells are a shared cultural object among many different Indigenous groups and are known from the archaeological and ethnohistoric records [e.g., [5][6][7][8][9][10][11][12][13]. More recently, turtle shells have sometimes been replaced with modern materials such as tin cans, wood, or copper, though the function is the same [14][15][16][17]. ...
... The second type, single shell design, is crafted similarly to the turtle shell leggings, except the turtle shells are not attached to hide. Instead, individual shells are tied to the arms or legs [8,26]. Native Americans may have preferentially selected box turtles due to the structure of their shells. ...
... Cooking or processing would cause the shell to be in at least three pieces (one carapace and two plastron pieces). The consumption of box turtles could lead to sickness, given that box turtles consume mushrooms that are poisonous to humans [8,[28][29]. The boiling process could also damage or weaken the shell, particularly along the sutures. ...
Article
Full-text available
Turtle shell rattles are percussion instruments used by Indigenous peoples of the Americas in ceremonial contexts to keep rhythm. Archaeological investigations in the southeastern United States produced several complete and partial Eastern box turtle (Terrapene carolina) shell rattles from mortuary contexts dating from the Archaic (ca. 8000–1000 BC) through Mississippian periods (ca. AD 800–1500). Fragmentary turtle remains, some identified as Eastern box turtle, are frequently recovered from non-mortuary contexts. Traditionally, these fragmentary remains are attributed to food waste. Given the archaeological and ethnographic evidence for turtle shell rattles, we need to consider how fragmentary remains might fit into the chaîne opératoire of rattle production. This paper presents the results of an experimental study designed to identify one such chaîne opératoire of rattle production. During this experiment, the data on taphonomic processes such as manufacturing marks, use-wear, and breakage patterns, were recorded. We then tested the taphonomic findings from the experimental study and an object trait list we compiled from known rattle specimens and documentary sources with archaeological turtle remains recovered from non-mortuary contexts at two Mississippian period (ca. AD 1000–1450) sites in Middle Tennessee. Historic indigenous groups are known to have, and still do into the present-day, make and use turtle shell rattles in the region. Ultimately, we determined that “food refuse” should not be the default interpretation of fragmentary box turtle remains, and instead the taphonomic history and contextual associations must be considered in full. The experimental process of crafting turtle shell rattles enhances our understanding of an ancient musical instrument and the success rate of identifying musical artifacts and distinguishing between other modified turtle remains in the archaeological record. This study expands our knowledge of ancient music in North America and prompts re-analysis of curated turtle remains in museums for rattle-related modifications.
... Since turtle shell rattles are generally associated with ritual, ceremonial, and mortuary activity, proper identification is important for indigenous groups, who may be able to claim cultural items under the Native Ameri-Turtle or tortoise (Testudines) shell rattles are percussion instruments used by indigenous peoples in ceremonial contexts to keep rhythm and are symbols of group beliefs (Jackson and Levine 2002; Figure 1). Turtle or tortoise shell rattles occur throughout the United States, ranging from California to New York to Florida (Brown 2011). However, in this article, we concentrate on turtle shell rattles in the southeastern United States, and particularly on the difficulty of identifying turtle shell rattles in the archaeological record. ...
... Through a review of several ethnographic and archaeological cases, we developed an object trait list of five common characteristics of rattle manufacturing that can be used to distinguish turtle shell rattles from subsistence remains, including: ceremonial, ritual, or burial association; turtle species preference; carapace and plastron representation; modifications; and rattle implements (Brown 2011). Rattle forms in the southeastern United States include parallel handheld and body (or shackle) types ( Figure 1). ...
... Shawnee, Cherokee, and Absentee of Oklahoma used Terrapene spp. to create body rattles or shackles, which are used by younger women, as opposed to the handheld rattles that are used by mostly men (Brown 2011;Jackson and Levine 2002;Voegelin 1942; Figure 1B-1C). The body rattle has been associated with various ceremonies and dances, such as the Garfish Dance, Green Corn Ceremony, Ribbon Dance, and the Stomp Dance (Dodd 2002:14;Howard 1968:90;Jackson and Levine 2002). ...
Article
Full-text available
The construction of rattles from turtle (Testudines) shells is an important consideration when distinguishing between food and non-food uses of archaeological turtle remains. However, the identification of turtle shell rattles in prehistoric contexts can be quite challenging. Equifinality is a major problem for being able to distinguish rattles from food refuse, particularly when a carapace is not burnt or modified. In addition, diversity, abundance, and distribution of Chelonian taxa varies throughout the southeastern United States, creating differential access for indigenous groups. Thus, multiple lines of evidence are needed from archaeological, ethnographic, and ethnohistoric records to successfully argue for the production and use of turtle shell rattles in the prehistoric southeastern United States. In this article, we present examples of turtle shell rattles in the southeastern United States to highlight their function and use by indigenous groups, the construction process, and several common characteristics, or an object trait list, that can aid in the identification of fragmentary turtle shell rattle remains. Proper identification of turtle remains is important for interpreting faunal remains and may be of interest to indigenous groups claiming cultural items under the Native American Graves Protection and Repatriation Act (NAGPRA).
... Tortugas marinas y seres humanos han establecido relaciones estrechas en la cohabitación de espacios litorales a lo largo de la historia y en diferentes partes del mundo (Ayres 1979, Allen 2007, Frazier et al. 2007, Woodrom 2010. Por lo general, las relaciones no se limitan a compartir un medio de vida, sino que es habitual que los seres humanos capturen tortugas para su alimentación (Gourou 1964, Frazier 1980, Woodrom 2003) y el aprovechamiento de huesos y caparazones como insumos en la manufactura de artefactos de diversa índole, tanto para fines prácticos, cotidianos y mundanos como para rituales simbólicos y ornamentales , Brown 2011, Frazier & Ishihara-Brito 2012, Gillreath-Brown & Peres 2017, 2018. Estos usos han sido documentados en sitios arqueológicos de distinta antigüedad en la península arábiga, en torno al océano Índico, en el Caribe centroamericano, el Yucatán, Norteamérica, la Polinesia y Sudamérica, demostrando su amplitud geográfica y profundidad histórica (Frazier 2005). ...
Article
Full-text available
La arqueología del litoral del desierto de Atacama ha con-centrado sus estudios en un espectro reducido de animales, sobre representando los más tradicionales, conocidos y populares, como peces, camélidos y otáridos, en desmedro de otros con un registro más acotado pero igualmente significativo. Presentamos una reflexión acerca de las relaciones entre seres humanos y tortugas marinas, destacando la diversidad de estrategias, instancias y mecanismos a través de los cuales estas relaciones se construían y actuaban.
... While periodic human burials appear at different types of Archaic sites within the Central Basin, major Archaic mortuary activity seems to be limited to large seasonal base camps. 1 These locations may represent areas where groups congregated periodically to bury their dead and exchange goods, given the presence of craft items such as modified turtle shell and caches of nonlocal stone tools (Brown 2011;Cridlebaugh 2017Cridlebaugh [1986; DeterWolf 2004;Gillreath-Brown and Peres 2017). Dense mortuary components at these sites may have acted as ancestral monuments or territorial markers, associating the sites and their surrounding territory with specific groups or lineages (Claassen 1996(Claassen , 2010Deter-Wolf and Peres 2014). ...
Chapter
Full-text available
Spatial analysis provides a greater understanding of relationships between people and environment. This chapter discusses settlement patterns and ecology of the Middle Cumberland River Valley (MCRV) of Tennessee through spatial analysis and a settlement ecological approach. Multiple lines of evidence are necessary to distinguish between survey bias and archaeological patterns. This chapter examines Archaic settlements at local and regional scales using GIS-based analyses. This chapter explores the environments that people chose to settle and the relationship to horticulture and gathering. The lower number of sites during the Middle Archaic could suggest a return to a residential mobility strategy, while the Late Archaic yielded more logistical mobility. The decrease in site numbers and elevation from the Early to Middle Archaic could be tied to temperature fluctuations during the Altithermal. Some Late Archaic sites occur at slightly higher elevations, indicating perhaps more varied resources. This chapter highlights Archaic patterns in MCRV with consideration of data limitations.
... The garfish is represented by a single scale. Both taxa are known to have strong ritual associations from other sites during the Archaic and/or Mississippian periods in Tennessee (Brown, 2011;Peres and Deter-Wolf, 2016a). ...
Article
Ancient Native American use of caves in the Eastern Woodlands occurred throughout the entire span of regional prehistory; however, the ways that these natural features were used varied considerably over time. To date only 25 cave sites containing deposits dated to the Archaic period (ca. 10,000-3000 B.P.) are recorded in the state of Tennessee, representing just 0.4% of the total known Archaic sites. In 2014 the authors conducted a salvage operation, bucket auger survey, and limited testing at the site of Black Cat Cave (40RD299) in Rutherford County, Tennessee to assess looting damage and assist in the installation of a security gate across the cave entrance. These investigations identified Black Cat Cave as the site of significant mortuary activity during the Middle Archaic (ca. 6460-6360 B.P.), and resulted in the recovery of rare Archaic faunal data from a cave setting. Analysis of faunal materials from the site allows us to add important new information to our understanding of ancient Native American landscape use in the Eastern Woodlands during the mid-Holocene.
... Moreover, most zooarchaeological work on tortoises conducted in different regions of the world (mainly from Spain, Italy, South Africa, Israel, Colombia, Caribbean islands, North America, etc.) are part of studies on diet amplitude, paleodemography, taphonomy, environmental changes, taxonomy and other non-food uses (e.g. Sampson, 1998Sampson, , 2000Klein and Cruz Uribe, 2000;Stiner et al., 2000;Speth and Tchernov, 2002;Frazier, 2005;Stahl and Oyuela-Caycedo, 2007;Blasco, 2008;Morales P erez and Sanchis Serra, 2009;Blasco et al., 2011;Brown, 2011;Salazar Garcia et al., 2013;Thompson and Henshilwood, 2014, among others). The researches analyzed not only different regions, but also various cultural contexts and timelines (from the Early Pleistocene to historical moments and ethnographic studies). ...
Article
Full-text available
Zooarchaeological analysis about role of turtles as resource of ancient inhabitants have been approach in different regions of the world, aboard the Broad-Spectrum Revolution and paleodemography frameworks and covering extensive periods of time. In the present work the role of tortoises (Chelonoidis chilensis) is discussed for the Beltrán Onofre Banegas-Lami Hernández site from Chaco-Santiagueña archaeological region (Santiago del Estero Province, Argentina). This site corresponds to the late agro-pottery stage, of short occupation timeline compared to works that treat this subject in the world. The proportion of used resources at the site was estimated and the ethnographic work of the study region and surrounding areas were analyzed to discuss importance of turtles in the diet of ancient inhabitants. In accord with the results C. chilensis was used opportunistically or circumstantially, and it could be more relevant during summer as additional resource next to lizards Tupinambis sp.
Technical Report
Citation: Cridlebaugh, Patricia A. 2017 Penitentiary Branch: A Late Archaic Cumberland River Shell Midden in Middle Tennessee. 2nd ed. edited by Aaron Deter-Wolf. Report of Investigations No. 4. Tennessee Department of Environment and Conservation, Division of Archaeology, Nashville.
Article
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
Chapter
The Zebree site is located in Mississippi County, Arkansas within the boundaries of the Big Lake National Wildlife Refuge. It is on an area of relatively high ground near the western edge of Big Lake itself, close to the Missouri state line (Fig. 11). The site is located on an old surface of an earlier braided Mississippi River stage. Typical lowland topography, including some prairie soils, the recent levee soils of Little River, and a large shallow lake, form the immediate environment. The site, numbered 3MS20, was one of the first sites mapped by the Arkansas State University research station after the founding of the Arkansas Archeological Survey in 1967. Four seasons of excavation were subsequently conducted at the Zebree site, each with different research goals (Thble 7; Anderson 1979). Three archaeological sites in the wildlife refuge were tested by the Arkansas Archeological Survey in 1968 (Morse 1968) because a drainage ditch was being planned through the refuge. Dan Morse and a crew of two placed two test pits in an area of old potholes at Zebree. Three separate components were recognized at the site, showing occupation during Late Woodland, early period Mississippian and middle period Mississippian times. Early period Mississippian was previously unidentified in Arkansas and was recognized as highly significant. This component was stratigraphically distinct from the Late Woodland and middle period Mississippian components. The discovery of significant stratified deposits at Zebree led to the funding by the National Park Service (Contract No. 14-10-7:911-21) of a major excavation at the site in 1969 (Morse 1975). Large contiguous areas of the site were dug in order to collect data on house patterns and community plans. Adequate and representative samples of artifacts from all three components were sought. Zooarchaeological samples were collected and identified. Hypotheses as diverse as the manufacture and function of the Zebree microlith industry, the date of the formation of Big Lake, and the nature and cause of cultural change to a Mississippian way of life were developed using data from this excavation. In particular, it was evident that the early period Mississippian Big Lake phase could not have developed from the preceding Late Woodland Dunklin phase (Morse 1977). Cultural processes including migration were examined to account for this development. Final construction plans for a ditch to prevent the silting of Big Lake made funding available for one more field season at the Zebree site (Morse and Morse 1980). The U.S. Army Corps of Engineers and the Arkansas Archeological Survey cooperated to mitigate the destruction of the Zebree site (Contract No. DACW 66-76-C-0006). The previous excavations provided sufficient information for detailed data recovery strategies such as a 1 percent stratified systematic unaligned sample of I-meter random squares, backhoe transects at judiciously chosen areas, and large area block excavations (Anderson 1976). Specialists hired to assist in recovering their own data in the field included an ethnobotanist, zoo archaeologist, and ceramic technologist. All samples were waterscreened. using mesh sizes from 1I16 to 1I4 inch. Flotation samples were automatically taken from each level and feature. Interaction between the nearby processing laboratory and field personnel was encouraged to maximize the range of samples recovered. A computer-generated map of the site was made of the different occupations using the randomsquare data. The botanical and zoological samples were studied to find both similarities and differences in diet of the three major prehistoric components. nee-ring coring and cores taken from the lake beds were both used to increase environmental data (Bowers 1976. King 1980). The historic nineteenth- and twentiethcentury components were also studied (P. Morse 1980a. 1980b). Two peer reviewers were flown in to review the excavations and make suggestions about data recovery (Smith 1976; Peebles 1976). A final report to the Corps was accepted in 1980 (Morse and Morse 1980). An unexpected opportunity to add to the already re covered data was presented to the excavators in 1976. The ditch did not go through the site until July. An enlightened contractor, S. J. Cohen of Blytheville, Arkansas, loaned the use of heavy machinery to a group of volunteers headed by Morse and Anderson. The overburden was cleared and defined features were uncovered and excavated. This emergency salvage added to the artifact inventory and knowledge of community planning. The new ditch and levee system was mapped in relationship to the remnant of the Zebree site that still remains.
Article
Anetso, a centuries-old Cherokee ball game still played today, is a vigorous, sometimes violent activity that rewards speed, strength, and agility. At the same time, it is the focus of several linked ritual activities. Is it a sport? Is it a religious ritual? Could it possibly be both? Why has it lasted so long, surviving through centuries of upheaval and change? Based on his work in the field and in the archives, Michael J. Zogry argues that members of the Eastern Band of Cherokee Nation continue to perform selected aspects of their cultural identity by engaging in anetso, itself the hub of an extended ceremonial complex, or cycle. A precursor to lacrosse, anetso appears in all manner of Cherokee cultural narratives and has figured prominently in the written accounts of non-Cherokee observers for almost three hundred years. The anetso ceremonial complex incorporates a variety of activities which, taken together, complicate standard scholarly distinctions such as game versus ritual, public display versus private performance, and tradition versus innovation. Zogry's examination provides a striking opportunity for rethinking the understanding of ritual and performance as well as their relationship to cultural identity. It also offers a sharp reappraisal of scholarly discourse on the Cherokee religious system, with particular focus on the Eastern Band of Cherokee Nation. © 2010 The University of North Carolina Press. All rights reserved.
Article
A major reconstruction of the rituals, cosmology, ideology, and political structures of the prehistoric native peoples of the Mississippi River Valley and Southeastern United States.