Content uploaded by Thomas Pluckhahn
Author content
All content in this area was uploaded by Thomas Pluckhahn on Dec 10, 2017
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ysea20
Download by: [154.16.6.163] Date: 09 December 2017, At: 18:55
Southeastern Archaeology
ISSN: 0734-578X (Print) 2168-4723 (Online) Journal homepage: http://www.tandfonline.com/loi/ysea20
Constructing community at civic-ceremonial
centers: pottery-making practices at Crystal River
and Roberts Island
Thomas J. Pluckhahn, Rachel E. Thompson & Kassie Kemp
To cite this article: Thomas J. Pluckhahn, Rachel E. Thompson & Kassie Kemp (2017)
Constructing community at civic-ceremonial centers: pottery-making practices at Crystal River and
Roberts Island, Southeastern Archaeology, 36:2, 110-121, DOI: 10.1080/0734578X.2016.1252190
To link to this article: https://doi.org/10.1080/0734578X.2016.1252190
Published online: 16 Nov 2016.
Submit your article to this journal
Article views: 174
View related articles
View Crossmark data
Citing articles: 2 View citing articles
Constructing community at civic-ceremonial centers: pottery-making practices at
Crystal River and Roberts Island
Thomas J. Pluckhahn
a
, Rachel E. Thompson
b
and Kassie Kemp
c
a
Department of Anthropology, University of South Florida, Tampa, FL, USA;
b
Florida Master Site File, Florida Division of Historical Resources,
Tallahassee, FL, USA;
c
Florida Public Archaeology Network–West Central Regional Center, Tampa, FL, USA
ABSTRACT
Crystal River (8CI1) and Roberts Island (8CI36, 8CI40, 8CI41) are neighboring mound complexes on
Florida’s west-central Gulf Coast, with mainly sequential occupations during the Middle and Late
Woodland periods, respectively. Previous work at Crystal River produced assemblages marked by
a diversity of pastes and surface treatments, suggestive of distinct communities of practice.
However, these excavations were unsystematic and poorly controlled, thus confounding
understanding of temporal and spatial variation in practice. Recent excavations in domestic
areas, combined with the analysis of older collections from mounds, support a finer-grained
understanding of variation in ceramic production. Our analysis suggests that communities of
practice persisted through time, although there is variation that corresponds well with changes
in settlement.
ARTICLE HISTORY
Received 2 June 2016
Accepted 20 October 2016
KEYWORDS
Communities of practice;
ceramics; ethnicity;
Woodland period
In recent years, southeastern archaeologists have come to
the realization that several of the largest civic-ceremonial
centers in the region did not simply develop in situ
among local populations, but instead formed through
the aggregation of people from diverse traditions. Per-
haps the best documented example is Cahokia, now
widely understood to have developed in large part
through the “centripetal immigration”(Emerson and
Hargrave 2000:4) of individuals and small groups of
varying ethnicities and from far-flung areas (Alt 2002,
2006,2012; Pauketat 2003,2007; Pauketat and Lopinot
1997). While the process of Mississippianization that
began at Cahokia was no doubt in many ways unique,
there may be commonalities to the broader outlines by
which civic-ceremonial centers develop, and not only
for the Mississippian period. For example, Sassaman
(2005:356) argues that the Late Archaic-period center
at Poverty Point formed through “…the syncretism of
two or more distinct ethnic groups.”
Emerson and Hargrave (2000:2) observe that the key
to understanding the social diversity inherent in the
development of civic-ceremonial centers, such as Caho-
kia, is the accumulation of “detailed historical and cul-
tural contextual analyses that distinguish constellations
of situationally significant materials and attributes …”
(see also Emerson and McElrath 2001:202). Unfortu-
nately, such detail is often lacking for many of the largest
civic-ceremonial centers in the region. One frequent
problem is a paucity of reliable radiocarbon dates.
Another problem is the longstanding –and in many
cases continuing –reliance on traditional ceramic typol-
ogies, which tend to reduce variation by relegating sherds
and vessels to a limited number of discrete types (Emer-
son 1999).
We describe ceramic assemblages from the Crystal
River site (8CI1) and the Roberts Island Shell Mound
Complex (principally sites 8CI36, 8CI40, and 8CI41),
neighboring Middle and Late Woodland-period civic-
ceremonial centers on the west-central Gulf Coast of Flor-
ida (Figure 1). The former site is famous as the southern-
most major expression of the Hopewell Interaction
Sphere, thanks to investigations of its Main Burial Com-
plex (Figure 2) by Moore (1903,1907,1918) and Bullen
(1951,1953,1966) in the early and middle twentieth cen-
tury, respectively. The Main Burial Complex includes a
dome-shaped mound (F) surrounded by a low platform
(E) and a circular embankment (C), all of which contained
burials. In addition to this mortuary complex, Crystal
River includes a separate burial mound (G), a plaza,
three platform mounds (A, H, and K), and an extensive,
deep comma-shaped midden. Roberts Island, which con-
sists of a complex of closely related sites (Weisman
1995b), is lesser known and, until recently only minimally
investigated. The complex includes three platform
mounds (A, B, and C) arranged around an apparent
plaza, with an extensive shell midden (Figure 3).
© Southeastern Archaeological Conference 2016
CONTACT Thomas J. Pluckhahn tpluckhahn@usf.edu Department of Anthropology, University of South Florida, 4202 E. Fowler Ave, SOC107, Tampa, FL,
USA
SOUTHEASTERN ARCHAEOLOGY, 2017
VOL. 36, NO. 2, 110–121
http://dx.doi.org/10.1080/0734578X.2016.1252190
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Previous reports of the pottery assemblages from
Crystal River suggest a diversity of formal types, includ-
ing: sand-tempered Deptford Check Stamped; sand-tem-
pered Swift Creek Complicated Stamped; various incised,
punctated, and red filmed types of the sand-tempered
Weeden Island series; plain and occasional other var-
ieties of the limestone-tempered Pasco series; and,
finally, plain and occasional other types of the St.
Johns series produced from clays heavily laden with
microscopic sponge spicules or diatomaceous earth,
but no macroscopically observable temper (Weisman
1995a). The diversity of tempers, in particular, would
seem plausibly explained by the coming together of
people with different pottery traditions, perhaps akin
to the formation of Cahokia or Poverty Point. Prior exca-
vations at Crystal River were unsystematic and poorly
controlled, however, and the analysis relied heavily on
normative assumptions about types, thus confounding
understanding of temporal and spatial variation in pot-
tery-making practice. Recent excavations in domestic
areas at Crystal River and Roberts Island, combined
with the analysis of older collections from mounds, sup-
port a finer-grained understanding of variation in cer-
amic production.
We employ a “communities of practice”perspective
(Lave and Wenger 1991) to understanding temporal
variation in pottery at Crystal River and Roberts Island.
As defined by Lave and Wenger (1991:98) a “commu-
nity of practice is a set of relations among persons,
activity, and world, over time and in relation with
other tangential and overlapping communities of prac-
tice.”Communities of practice theory focuses on how
an individual is integrated into a group by “peripheral
learning,”through imitation, observation, and gradu-
ated participation (Lave and Wenger 1991:29; Wenger
1998). Originally developed under the disciplines of
education and psychology, “situated learning”theory
and “legitimate peripheral learning”were ways to
understand knowledge transference (Lave and Wenger
1991:29–31). Lave and Wenger’s(1991:30–31) study
of crafting in West Africa led them to the conclusion
that situated learning is not simply hands-on, but a
vital part of social practice. Learning the ways of the
community is part of becoming a member of that com-
munity and its social circle. In the sense of communities
of practice, “learning is not merely a condition for
membership, but is itself an evolving form of member-
ship”(Lave and Wenger 1991:53).
Figure 1. Location of the Crystal River site and Roberts Island complex.
SOUTHEASTERN ARCHAEOLOGY 111
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Situated learning and communities of practice pro-
vide a useful framework for understanding temporal
changes in ceramics at Crystal River and Roberts Island
and, by extension, the scales of social interaction and
social diversity through time. All stages of the pottery
chaîne opératoires can reflect community membership.
Of course, as Gosselain (2000:190) states, the trans-
mission of ideas is not the same for all the stages of pot-
tery production. Some stages are arguably more easily
affected by subtle changes, such as the mimicking of dec-
oration or clay selection, while other production stages
such as paste creation and vessel forming require a
much more substantial influence to alter technique (Gos-
selain 2000:191–192; Lyons and Clark 2012:24–25;
Minar 2001; van der Leeuw 1984,1993).
Communities of practice theory is particularly useful
for understanding possible changes in learning associ-
ated with ceramic production in archaeological contexts,
in that it does not require that we specify the precise
relationship of the transfer of knowledge (Wendrich
2012a). It thus avoids the difficult issue of identifying
ethnicity in the archaeological record (Jones 1997; Shen-
nan 1989), a possibility we do not exclude but also do not
assume. Additionally, this perspective leaves open the
possibility that individual potters may have been
enmeshed in multiple communities of practice (Wen-
drich 2012b).
Methods
The midden assemblages under consideration derive
from recent work at the sites under the auspices of the
Crystal River Early Village Archaeological Project (CRE-
VAP), a study funded by the National Science Foun-
dation under the direction of Pluckhahn, Victor
Thompson, and Brent Weisman. Based on Bayesian
modeling of more than 30 recently obtained radiocarbon
dates on stratified midden layers and features, Pluckhahn
et al. (2015b) (see also Cherkinsky et al. 2014) divide the
domestic occupation of Crystal River and Roberts Island
into four phases. Thompson (2016) used these phases to
examine change in domestic pottery assemblages over
time, focusing on pottery from well-dated midden con-
texts. We follow the same tactic here. These well-dated
contexts are admittedly somewhat limited in terms of
spatial coverage, including selected levels from four 1-
×-1-m test units at Crystal River (see Figure 2), and
three 50-×-50-cm shovel tests and one 1-×-2-m test
trench at Roberts Island (see Figure 3). As an additional
Figure 2. Map of the Crystal River site.
Figure 3. Map of the Roberts Island complex.
112 T. J. PLUCKHAHN ET AL.
Downloaded by [154.16.6.163] at 18:55 09 December 2017
caveat, because of variations in the duration, intensity,
and extent of occupation, the assemblages from some
phases are smaller than others.
The mound assemblages described here result from
the previous excavations by Bullen (1951,1953,1966),
as well as our own, much more modest sampling. We
rely primarily on Kemp’s(2015) re-analysis of the Crys-
tal River mound assemblages derived from previous
excavations by Bullen, curated at the Florida Museum
of Natural History. Pluckhahn has also examined the
ceramics recovered by Moore, now curated at the
National Museum of the American Indian, which we
address in qualitative terms. For the mounds at Roberts
Island, we focus on the ceramics recovered from a 1-×-6-
m trench on the slope of Mound A, a 1-×-4-m trench on
the slope of Mound B, and a 50-×-50-cm shovel test on
the summit of Mound C.
Bullen’s mound excavations are poorly documented,
which limits their utility for understanding change
through time. However, we conducted Geoprobe cor-
ing and geophysical survey of all but the burial
mounds to gain greater insight into the nature and
timing of mound construction. For the burial mounds,
we also have retrieved new dates on artifacts and
human remains recovered by Bullen. Discounting a
few early dates retrieved by Bullen that have error
ranges of more than 100 years, there are presently 18
radiocarbon dates from six mounds or mound com-
plexes at Crystal River; this includes three retrieved
dates by Katzmarzyk (1998:Tables 3–8, 3–9) and 15
recovered by Pluckhahn and colleagues (2015a; see
also Norman 2014). For Roberts Island, we have one
AMS date for two of the three mounds. The timing
for Mounds A and H at Crystal River and Mound A
at Roberts Island benefit further from optically stimu-
lated luminescence (OSL) dating of sediments within
and below the monuments (Pluckhahn et al. 2015a).
As with the midden dates, we conducted Bayesian
modeling of the start and end dates for mound con-
struction in OxCal 4.2 (
©
Christopher Bronk Ramsey
2013; Bronk Ramsey 2009).
A few cautions should be borne in mind with regard
to our discussion of mound assemblages and their tim-
ing. First, because of the absence of detailed notes or
maps, Bullen’s mound assemblages can only be tied to
general proveniences. Second, our reconstruction of
mound construction episodes is often based on the
very limited windows on stratigraphy provided by
small diameter cores and geophysics. We omit Mound
J from consideration because its odd shape and stratigra-
phy make it difficult to interpret. Finally, we note that the
ceramic assemblages from mounds do not necessarily
reflect the period of use.
The pottery from both midden and mounds was
sorted by temper and surface attributes. We used pre-
viously defined types (e.g., Goggin 1952; Willey 1949)
where appropriate, but we also were attuned to variations
that could not be subsumed within these discrete cat-
egories. For example, our analysis suggested that a sig-
nificant fraction of the pottery at Crystal River and
Roberts Island contains a mix of aplastics –especially
sand and limestone –that might suggest mingled pro-
duction practices. Although our analysis also extended
to vessel form and use-wear (Kemp 2015; Thompson
2016), we omit these in the interest of space. Since the
majority of the vessels from domestic assemblages
from all phases are simple open bowls, however, these
analyses would have added little more to our
interpretation.
Results
We structure the discussion by phase. Table 1 summar-
izes the pottery assemblages from middens by phase. The
pottery assemblages from the Crystal River and Roberts
Island mounds are summarized in Tables 2 and 3.
Phase 1
Phase 1 of midden deposition has a modeled start date of
cal A.D. 69 to 225 and an end date of cal A.D. 144 to 265
(here and elsewhere, we focus on the 95 percent prob-
abilities of the posterior density estimates, and follow
the convention of using italics to distinguish Bayesian
modeled dates (Bayliss et al. 2011; Pluckhahn et al.
2015b:31). The assemblage from well-dated midden con-
texts for this phase is limited to only 37 sherds from the
lowermost levels in two test trenches at Crystal River, so
our conclusions are necessarily limited.
The Phase 1 assemblage consists of an almost even
mix of sand and limestone-tempered pottery, with few
mixed-temper sherds, and St. Johns-like wares. Surface
treatments consist only of plain and check stamped. If
we assume the pottery from this phase is locally made,
given the lack of unusual tempers or surface treatments,
it would be consistent with the emergence of Crystal
River as a ceremonial center for a community dispersed
across the local landscape, perhaps especially on the
marsh islands to the west. Based on its limited spatial
extent, the Phase 1 resident population at Crystal River
appears to have remained relatively small (Pluckhahn
et al 2015b). Isotopic studies of oysters suggest it may
have been restricted to a seasonal occupation during
colder months, perhaps in association with ceremonies
that included mound construction (Thompson et al.
2015).
SOUTHEASTERN ARCHAEOLOGY 113
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Although we suggest that the seasonal inhabitants of
Crystal River were drawn from the local area, the almost
equal division between sand and limestone tempering in
the Phase 1 midden assemblage suggests that it neverthe-
less may have included two distinct communities of pot-
tery-making practice. Looking at sherd cross-sections
under magnification, the extremes in tempering are
striking: some sherds are densely packed with fine, med-
ium, and coarse light-colored nodules (mainly lime-
stone), others lack these large inclusions entirely but
are very sandy, and still others have no apparent temper
or inclusions. Crystal River is situated on one of many
limestone outcrops in this karst landscape. In core
samples, we noted sandy clays lying just above the lime-
stone substrate (see Norman 2014) that would seem to
have been suitable for pottery making. Ann Cordell
(2016) conducted ceramic petrography of a sample
from one such on-site clay layer, as well as three other
clays from exposed beds on nearby islands and main-
land. Two of the samples had high sand content and
two moderate (with one of these also including sponge
spicules). None of the clays contained limestone,
although nodules of gypsum and phosphate were noted.
Thus it would seem that the people living at Crystal
River had a choice with regard to clay recipes –they
had easy access to relatively pure clays and also to lime-
stone that could be used for temper. Limestone outcrops
are rare with increasing distance from the mainland, and
we might therefore expect this temper to have been less
commonly utilized by people who resided on the marsh
islands downstream and seaward of Crystal River.
The possible social division that is evident in temper
may be repeated in the mounds. Radiocarbon dating
suggests that Mound G may have been one of the earliest
architectural features initiated at Crystal River, perhaps
even beginning a century or two before the first domestic
occupation (Katzmarzyk 1998; Milanich 1999:23). The
identifiable sherds from this mound total only 155
(Kemp 2015). Generally consistent with the Phase 1-
midden assemblage, the majority (70 percent) of the
sherds from Mound G are limestone-tempered, with
sand-tempered following at 18 percent. Also consistent
with the Phase 1-midden assemblage, most of the sherds
from Mound G are plain, with the few decorated types
limited to a few sherds at most.
The Main Burial Complex also appears to have
formed early in the history of Crystal River, and, like
Mound G, may have been initiated even before the vil-
lage was occupied. The ceramic assemblages from the
complex are clearly different from those from Mound
Table 1. Summary of ceramics from well-dated midden contexts by phase (from Thompson 2016).
Paste group Type/description Phase 1 midden Phase 2 midden Phase 3 midden Phase 4 midden
N(%) N(%) N(%) N(%)
Residual/unidentified 40 2377 360 960
Sand/grit tempered Franklin/Weeden Island Plain 10 (27.03) 286 (16.17) 113 (43.13) 269 (51.04)
Deptford Linear Check Stamped 7 (18.92)
Deptford Check Stamped 2 (0.11)
Deptford Simple Stamped 2 (0.11)
Wakulla Check Stamped 89 (33.97) 45 (8.54)
Swift Creek Complicated Stamped 7 (0.40) 13 (4.96) 2 (0.38)
unidentified stamped 5 (0.28) 2 (0.76) 12 (2.28)
Weeden Island Red 9 (0.51) 3 (0.57)
Weeden Island Zoned Punctate 4 (0.76)
Weeden Island Zoned Incised/Punctate 2 (0.11)
Ruskin Dentate Stamped 2 (0.38)
UID complicated stamped 2 (0.11)
Carrabelle Incised 2 (0.38)
UID incised 1 (0.06) 1 (0.38)
UID cord marked 1 (0.06)
Total sand/grit tempered 17 (45.95) 317 (17.92) 218 (83.21) 339 (64.33)
Limestone tempered Pasco Plain 18 (48.65) 1230 (69.53) 41 (15.65) 107 (20.30)
Pasco Check Stamped 1 (2.70)
Pasco Simple Stamped 1 (0.06)
Pasco Red 29 (1.64)
Total limestone tempered 19 (51.35) 1260 (71.23) 41 (15.65) 107 (20.30)
St. Johns like St. Johns Plain 17 (0.96) 2 (0.76) 5 (0.95)
Dunns Creek Red 1 (0.06)
St. Johns Check Stamped 1 (0.19)
St. Johns Incised 1 (0.19)
St. Johns Scored 1 (0.06)
Total St. Johns like 0 (0.00) 19 (1.07) 2 (0.76) 7 (1.33)
Other and mixed temper Plain, mixed sand/limestone 1 (2.70) 171 (9.67) 74 (14.04)
Plain, shell tempered 1 (0.06) 1 (0.38)
UID incised, mixed sand/limestone 1 (0.06)
Total other and mixed temper 1 (2.70) 173 (9.78) 1 (0.38) 74 (14.04)
Total identifiable 37 (100.00) 1769 (100.00) 262 (100.00) 527 (100.00)
114 T. J. PLUCKHAHN ET AL.
Downloaded by [154.16.6.163] at 18:55 09 December 2017
G and the midden (Kemp 2015). The 1942 total identifi-
able sherds from Mounds E and F are dominated by
sand-tempered and St. Johns-like ceramics, with less
amounts of limestone-tempered. The pattern is similar
for the 11,468 identifiable sherds from Mound
C. Surface decorations from the Main Burial Complex
Table 2. Summary of ceramics from Crystal River Mounds A-G (from Kemp 2015).
Paste group Type/description
Crystal River
Mound A
Crystal River
Mound C
Crystal River
Mounds E and F
Crystal River
Mound G
N(%) N(%) N(%) N(%)
Residual/unidentified 463 70 14
Sand/grit tempered Franklin/Weeden Island Plain 2 (4.88) 4644 (40.50) 467 (24.05) 24 (15.48)
Deptford Linear Check Stamped 47 (0.41) 5 (0.26)
Deptford Check Stamped 31 (0.27) 4 (0.21) 1 (0.65)
Deptford Simple Stamped 18 (0.16) 9 (0.46) 1 (0.65)
Wakulla Check Stamped 417 (3.64) 68 (3.50)
Swift Creek Complicated Stamped 82 (0.72) 30 (1.54) 1 (0.65)
Weeden Island Red 1024 (8.93) 94 (4.84) 1 (0.65)
Weeden Island Zoned Red 20 (0.17) 3 (0.15)
Weeden Island Zoned Incised 13 (0.11) 4 (0.21)
Weeden Island Zoned Punctate 87 (0.76) 42 (2.16)
Weeden Island Zoned Incised/Punctate 3 (0.03) 2 (0.10)
Ruskin Dentate Stamped 139 (1.21) 13 (0.67)
Ruskin Linear Punctated 6 (0.05)
Crystal River Negative Painted 1 (0.05)
New River Complicated Stamped 1 (0.01)
Crooked River Complicated Stamped 2 (0.02) 8 (0.41)
St Andrews Complicated Stamped 12 (0.10)
Tampa Complicated Stamped 23 (0.20)
Old Bay Complicated Stamped 1 (0.01)
UID stamped
Carrabelle Incised 23 (0.20)
Carrabelle Punctate 28 (0.24)
UID incised
UID punctate
Hillsborough Shell Stamped 26 (0.23)
Lochloosa Punctate 3 (0.03)
Indian Pass Incised 7 (0.06)
Keith Incised 23 (0.20) 8 (0.41)
Thomas Simple Stamped 13 (0.11) 3 (0.15)
West Florida Cord Marked 5 (0.04) 2 (0.10)
St Petersburg Incised 1 (0.01)
Safety Harbor Incised 1 (0.01)
Safety Harbor Punctate 3 (0.15)
Pinellas Plain 1 (0.05)
Total sand/grit tempered 2 (4.88) 6700 (58.42) 767 (39.50) 28 (18.06)
Limestone tempered Pasco Plain 35 (85.37) 1512 (13.18) 329 (16.94) 109 (70.32)
Pasco Check Stamped 23 (0.20)
Pasco Red 166 (1.45) 67 (3.45)
Pasco Punctated
Pasco Incised 5 (0.04) 2 (0.10)
Total limestone tempered 35 (85.37) 1706 (14.88) 398 (20.49) 109 (70.32)
St. Johns like St. Johns Plain 3 (7.32) 1958 (17.07) 565 (29.09) 1 (0.65)
Dunns Creek Red 905 (7.89) 183 (9.42)
St. Johns Check Stamped 1 (2.44) 3 (0.03) 1 (0.05) 3 (1.94)
St. Johns Simple Stamped 1 (0.05)
St. Johns Incised 5 (0.04) 4 (0.21)
St. Johns Punctated 1 (0.01)
Oklawaha Plain 1 (0.01)
Oklawaha Incised 6 (0.05)
Papys Bayou Plain 1 (0.01) 1 (0.05)
Papys Bayou Punctated 25 (0.22) 1 (0.05)
Papys Bayou Incised 5 (0.04)
Little Manatee Zoned Stamped 2 (0.02)
UID stamped
Total St. Johns like 4 (9.76) 2912 (25.39) 756 (38.93) 4 (2.58)
Other and mixed temper Plain, mixed sand/limestone 22 (0.19) 1 (0.05)
Plain, sand with UID inclusions 100 (0.87) 4 (0.21)
Plain, sand with mica 25 (0.22) 13 (0.67)
Plain, fullers earth? 3 (0.03) 3 (0.15)
Plain, shell tempered
Total other and mixed temper 0 (0.00) 150 (1.31) 21 (1.08) 0 (0.00)
Total identifiable 41 (100.00) 11468 (100.00) 1942 (100.00) 155 (100.00)
SOUTHEASTERN ARCHAEOLOGY 115
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Table 3. Summary of ceramics from Crystal River Mounds H and K, and Roberts Island Mound A–C (from Kemp 2015 and Thompson
2016).
Paste group Type/description
Crystal River
Mound H
Crystal River
Mound K
Roberts Island
Mound A
Roberts Island
Mound B
Roberts Island
Mound C
N(%) N(%) N(%) N(%) N(%)
Residual/unidentified 1 2 638 554 15
Sand/grit
tempered
Franklin/Weeden Island Plain 3 (10.71) 2 (6.25) 95 (47.26) 135 (48.56) 13 (81.25)
Deptford Linear Check
Stamped
Deptford Check Stamped
Deptford Simple Stamped
Wakulla Check Stamped 6 (2.99) 31 (11.15) 1 (6.25)
Swift Creek Complicated
Stamped
23 (8.27)
Weeden Island Red 3 (1.49)
Weeden Island Zoned Red
Weeden Island Zoned
Incised
Weeden Island Zoned
Punctate
1 (0.50)
Weeden Island Zoned
Incised/Punctate
Ruskin Dentate Stamped 1 (0.50) 5 (1.80)
Ruskin Linear Punctated
Crystal River Negative
Painted
New River Complicated
Stamped
Crooked River Complicated
Stamped
St Andrews Complicated
Stamped
Tampa Complicated
Stamped
Old Bay Complicated
Stamped
UID stamped 7 (3.48) 29 (10.43)
Carrabelle Incised
Carrabelle Punctate
UID incised 1 (0.50)
UID punctate 1 (0.36) 1 (6.25)
Hillsborough Shell Stamped
Lochloosa Punctate 1 (3.57)
Indian Pass Incised
Keith Incised
Thomas Simple Stamped
West Florida Cord Marked 1 (0.36)
St Petersburg Incised
Safety Harbor Incised
Safety Harbor Punctate
Pinellas Plain
Total sand/grit tempered 1 (3.57) 2 (6.25) 114 (56.72) 225 (80.94) 15 (93.75)
Limestone
tempered
Pasco Plain 22 (78.57) 22 (68.75) 57 (28.36) 5 (1.80)
Pasco Check Stamped
Pasco Red
Pasco Punctated 1 (0.36)
Pasco Incised
Total limestone tempered 22 (78.57) 22 (68.75) 57 (28.36) 6 (2.16) 0 (0.00)
St. Johns like St. Johns Plain 2 (7.14) 3 (9.38) 11 (5.47) 28 (10.07)
Dunns Creek Red 1 (0.50) 1 (0.36)
St. Johns Check Stamped 3 (1.49) 11 (3.96) 1 (6.25)
St. Johns Simple Stamped
St. Johns Incised
St. Johns Punctated 1 (0.50)
Oklawaha Plain
Oklawaha Incised
Papys Bayou Plain
Papys Bayou Punctated
Papys Bayou Incised
Little Manatee Zoned
Stamped
1 (0.36)
UID stamped 1 (0.36)
Total St. Johns like 2 (7.14) 3 (9.38) 16 (7.96) 42 (15.11) 1 (6.25)
(Continued)
116 T. J. PLUCKHAHN ET AL.
Downloaded by [154.16.6.163] at 18:55 09 December 2017
exhibit a much greater variety than is apparent in the
midden. Although the majority of the sherds are still
plain, Weeden Island Red and other types contribute
substantially to the overall assemblages from Mound C
and Mounds E and F, even if each individual type is
low in frequency.
These differences in the ceramic assemblages from the
Main Burial Complex relative to Mound G and Phase
1 midden contexts probably reflect a longer period of
use for the former mortuary feature. However, the pot-
tery assemblages also may reveal social differences,
given the placement of separate burial facilities on
opposing sides of the plaza. The discrepancy in burial
goods also is conspicuous; burials in Mound G produced
few grave goods, and those that were present were pri-
marily made from locally obtainable shell. Mound F in
the Main Burial Complex produced most of the Hope-
well exotics for which Crystal River is famous (Moore
1903,1907,1918). Presumably, many of the ceramics
from the Main Burial Complex were imported from
other regions; examination of Moore’s vessels from the
Main Burial Complex indicated that many were made
from micaceous clays that would not be found in penin-
sular Florida.
Phase 2
Phase 2 is the longest of the four intervals of occupation
at Crystal River, with a modeled start date of cal A.D.
221–321 and a modeled end date of cal A.D. 434–544
(Pluckhahn et al. 2015b:31). During this phase, the vil-
lage at Crystal River grew rapidly to its full areal extent.
It also grew vertically with the deposition of copious
quantities of midden comprised mainly of oyster shell.
Finally, the occupation appears to have become more
permanent; isotopic studies of oysters from this phase
exhibit evidence for collection during all seasons
(Thompson et al. 2015). Elsewhere (Pluckhahn and
Thompson 2017), we have speculated that the develop-
ment of a larger and more permanent village may have
been given impetus by a pulse in sea level rise in the
first centuries A.D. (Goodbred et al. 1998; McFadden
2015,2016) that made life on low-lying marsh islands
more tenuous, although we also suggest that the increas-
ing elaboration of the ceremony might have served as a
powerful attraction.
In the pottery assemblage from Phase 2 midden con-
texts, limestone tempering comes to dominate over sand,
forming over 70 percent of the assemblage compared to
less than 20 percent sand (Thompson 2016). This pattern
is consistent across various sampling proveniences, and
suggests the establishment of a more uniform, local pot-
ting tradition. Surface treatments also are consistent with
this; although a number of named types are present,
more than 96 percent of the midden pottery from
Phase 2 is plain.
A possible blending of pottery-making traditions or
communities is indicated by the observation that nearly
10 percent of sherds from Phase 2 midden contexts
appear to have a fabric comprised of mixed sand and
limestone. Such mixing may be much more common
than allowed for by our macroscopic sorting; gross
paste analysis has revealed that much of the pottery we
classified as sand contained small amounts of limestone,
and that many of the sherds we classified as limestone-
tempered contained sand (Kemp 2015; Thompson
2016).
Recent dating suggests that two of the smaller plat-
form mounds at Crystal River –Mounds H and K –
were constructed during Phase 2 (Norman 2014). The
pottery assemblages from these mounds, although lim-
ited in size, are remarkably similar to those from the
midden. The FLMNH has collections from two units
Bullen (1966) excavated on or near the summit of
Mound H, the small ramped platform mound anchoring
the north end of the plaza. Only 28 identifiable sherds are
present, but the proportions are remarkably similar to
the midden in regard to temper, with a majority of lime-
stone-tempered (Kemp 2015). Also like the Phase 2 mid-
den, 97 percent of the Mound H assemblage is plain.
Bullen (1966) excavated one unit in Mound K. The
assemblage of identifiable pottery from this mound is
Table 3. Continued.
Paste group Type/description
Crystal River
Mound H
Crystal River
Mound K
Roberts Island
Mound A
Roberts Island
Mound B
Roberts Island
Mound C
N(%) N(%) N(%) N(%) N(%)
Other and mixed
temper
Plain, mixed sand/limestone 2 (6.25) 13 (6.47) 5 (1.80)
Plain, sand with UID
inclusions
3 (9.38)
Plain, sand with mica
Plain, fullers earth?
Plain, shell tempered 1 (0.50)
Total other and mixed
temper
0 (0.00) 5 (15.63) 14 (6.97) 5 (1.80) 1 (6.25)
Total identifiable 28 (100.00) 32 (100.00) 201 (100.00) 278 (100.00) 16 (100.00)
SOUTHEASTERN ARCHAEOLOGY 117
Downloaded by [154.16.6.163] at 18:55 09 December 2017
small (n= 32), but again shows great consistency with
the Phase 2 midden in the dominance of limestone (at
65 percent) over other tempers, and plain (at 97 percent)
over other surface treatments (Kemp 2015). Also similar
to the Phase 2-midden ceramics, the Mound K assem-
blage exhibits a relatively high frequency (15 percent)
of mixed-temper sherds.
Phase 3
Phase 3 has a modeled start date of cal A.D. 478–634 and
an end date of cal A.D. 663–810 (Pluckhahn et al.
2015b:32). This phase saw a waning of settlement at
Crystal River; the village contracted to the area north
of Mound A and there was less midden deposition.
Some of the population may have moved to Roberts
Island, where settlement initiated during this interval.
Other people probably dispersed across similar small
settlements. The settlement shift at Crystal River was
mirrored at contemporaneous mound centers across
much of the Gulf Coast (Menz 2015; Russo et al. 2014;
Wallis and McFadden 2016; Wallis et al. 2015). The
causes of this apparent region-wide reorganization of
settlement are currently unclear, but shifts to a drier or
more variable (or both) climate have been implicated
for some of the changes in the region during this interval
(Marquardt 2014:10; Sassaman 2012:262; Smith 2009).
Pottery appears to track these changes in settlement,
although the precise mechanisms for this are not clear.
In the Phase 3 midden assemblages, sand (at 83 percent)
comes to dominate over limestone (16 percent), and
there is a reduction in the proportion of mixed-temper
sherds (Thompson 2016). The increase in sand at the
expense of limestone is reminiscent of Phase 1, and
suggests that this temper is more common in this area
when the population was more dispersed, rather than
concentrated at Crystal River. The lesser proportion of
mixed tempers could be consistent with potters in
greater isolation from each other.
The surface treatments that are represented in Phase 3
midden assemblages show greater spatial variability than
those of preceding phases. The sub-assemblage from
Crystal River is almost entirely plain, and thus shows
continuity with Phase 2. However, the Phase 3 sub-
assemblage from Roberts Island is more varied, with sig-
nificant amounts of Swift Creek Complicated Stamped
and Wakulla Check Stamped and trace amounts of sev-
eral other decorative types.
Recent work suggests that Mound A, the large plat-
form mound at Crystal River, was completed during
Phase 3, even as the settlement of the site was diminish-
ing. No excavations have been conducted in Mound A,
but Bullen (1951) and Smith (1951) surface collected a
minimal number of sherds. Kemp (2015) documented
41 sherds from these investigations. In contrast with
the midden, most are limestone-tempered. The assem-
blage is entirely plain, with the exception of one
St. Johns Check Stamped. The discrepancies between
the mound and midden assemblages suggest the possi-
bility that Mound A may have been constructed at
least partially of repurposed midden. This finds some
support in the stratigraphic inversion of carbon dates
(Norman 2014). It also may be supported by isotopic
studies of oysters (Thompson et al. 2015). Samples
from Mound A were determined to have been gathered
across seasons, as with those from the midden; in con-
trast, oyster samples from all of the other mounds that
were sampled were only harvested in cooler months.
Phase 4
The final phase of occupation at Crystal River and
Roberts Island has a modeled start date of cal A.D.
723–881 and end date of cal A.D. 891–1060 (Pluckhahn
et al. 2015b:31). By this point, Crystal River appears to
have been abandoned except for a small area of occu-
pation immediately north of Mound A. Conversely,
this was the period when most of the midden was depos-
ited at Roberts Island; radiocarbon dates indicate it also
was the interval when at least two of the three mounds
(Mounds A and B) were constructed (Mound C is
undated).
Sand-tempered pottery remained dominant in Phase
4 midden assemblages, although less so than in Phase 3
(Thompson 2016). The proportions of limestone-tem-
pered and St. Johns-like sherds increased slightly, while
mixed-tempered sherds increased more dramatically.
Surface decoration remained mostly plain but with con-
tinued trend toward a greater richness in minority wares,
which now include a variety of Weeden Island types,
St. Johns and Wakulla Check Stamped, and Swift
Creek Complicated Stamped. However, it should be
noted that there is considerable variation in the pro-
portions of both tempers and surface treatments across
proveniences. These trends and the diversity across
sampling units are difficult to interpret, but could be
consistent with more seasonal use of both settlements
in association with ceremonies.
The assemblages from the three mounds at Roberts
Island are generally consistent with those from the mid-
den in terms of the dominance of sand-tempered pot-
tery. However, the mounds show greater proportions
of St. Johns-like pottery than the midden, and lesser rela-
tive frequencies of mixed-temper sherds. As with the
diversity in the Phase 4 midden, this is difficult to
interpret.
118 T. J. PLUCKHAHN ET AL.
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Discussion
Potters at Crystal River made choices –probably both
deliberate and unconscious –regarding pastes and dec-
orations. These choices persisted across phases of occu-
pation and monument construction, but not without
potentially significant variation. The close correspon-
dence between temporal changes in the ceramic attri-
butes (especially temper) in midden assemblages and
shifts in settlement suggest that these changes may be
explained by variation in the intensity of interaction
among communities of potters. Such variation might
be expected with the waxing and waning of civic-cere-
monial centers.
For Phase 1, when Crystal River appears to have had a
relatively small and seasonal resident population, the
relatively even split between limestone- and sand-tem-
pered pottery in the midden suggests to us that it was
a ceremonial center for two distinct communities of pot-
ters. This interpretation is lent some additional credence
by the apparent maintenance of two distinct burial facili-
ties. Presumably, given the lack of exotic tempers or sur-
face decorations from midden contexts, these were social
groups from the general area. But it is clear that the site
also attracted people or materials, or both, from farther
afield, as evidenced by the diversity of ceramics and
other burial goods in the Main Burial Complex.
During Phase 2, the village at Crystal River grew sub-
stantially in both size and permanence, at a rate dramatic
enough to suppose that it must have involved an influx of
people. The pottery from midden assemblages suggest
both a convergence toward locally-available limestone
aplastics and a blending of ceramic-making practices as
potters came into more sustained contact, a pattern simi-
lar to those observed in the ceramics from similar
coalescent communities elsewhere (e.g., Eckert 2012;
Garraty 2013; Lyons and Clark 2012; Roddick 2009; Sas-
saman and Rudolphi 2001).
Phases 3 and 4 are somewhat more difficult to inter-
pret. In general, however, the higher relative frequencies
of sand-tempered pottery and greater variation in surface
decoration (even as plain continues to dominate) may be
consistent with a renewed dispersal of population to the
seaward marsh islands as Crystal River gradually
declined in population and ritual importance, and
Roberts Island grew.
As Lave and Wenger (1991) and others (e.g., Roberts
2006) have noted, communities of practice are not static.
It follows that our archaeological approaches to commu-
nities of practice must be able to recognize temporal
fluidity in material remains. This is perhaps especially
true for the communities that converged at civic-ceremo-
nial centers of the scale of Crystal River, which we now
know frequently involved the aggregation of people
from different areas and traditions of material culture.
Given the difficulties of definitively identifying ethni-
city from the material record (Jones 1997), we are not
ready to go so far as to necessarily equate differences
in pottery-making practice at Crystal River and Roberts
Island with distinct ethnic groups, as has been argued
from other facets of material culture for the civic-cere-
monial centers at Poverty Point (Sassaman 2005) and
Cahokia (Emerson and Hargrave 2000). As we have
noted, much of the variation in temper may be related
to shifts in settlement and concomitant variability in
the availability of particular clays and aplastics. Still, we
recognize ethnicity as a possible explanation for ceramic
variation and, by extension, social difference. Emerson
(1999) has cogently argued for the understanding of eth-
nicity not in terms of discrete and bounded social groups
but instead as a process, akin to what we suggested above
for communities of practice. To the extent that we can
extrapolate from communities of pottery-making prac-
tice to ethnic groups and identities, the evidence pre-
sented here would seem to suggest that the process was
uneven.
Acknowledgments
We thank the Florida Museum of Natural History for provid-
ing access to their collections. Ann Cordell, Donna Ruhl, and
Neill Wallis provided assistance to Kemp during her work at
the museum. For the archaeological work at Crystal River,
we thank Crystal River Archaeological State Park, the Florida
Department of Environmental Protection, and the Florida
Division of Historical Resources.
Data availability statement
Bullen’s collections from Crystal River are curated at the
Florida Museum of Natural History in Gainesville. The
artifacts from CREVAP excavations at Crystal River
and Roberts Island are temporarily curated at the Uni-
versity of South Florida Department of Anthropology,
with final curation at the Florida Bureau of Archaeologi-
cal Research.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
This material is based upon work supported by the National
Science Foundation under Grant No. 1026248. Any opinions,
findings, and conclusions or recommendations expressed in
this material are those of the author(s) and do not necessarily
reflect the views of the National Science Foundation.
SOUTHEASTERN ARCHAEOLOGY 119
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Notes on contributors
Thomas J. Pluckhahn, Department of Anthropology, Univer-
sity of South Florida, Tampa, FL, USA.
Rachel E. Thompson, Florida Master Site File, Florida Division
of Historical Resources, Tallahassee, FL, USA.
Kassie Kemp, West-Central Regional Center, Florida Public
Archaeology Network, Tampa, FL, USA.
References
Alt, Susan M. 2002 Identities, Traditions, and Diversity in
Cahokia’s Uplands. Midcontinental Journal of Archaeology
27:217–236.
Alt, Susan M. 2006 The Power of Diversity: The Roles of
Migration and Hybridity in Culture Change. In
Leadership and Polity in Mississippian Society, edited by B.
M. Butler and P. D. Welch, pp. 289–308. Center for
Archaeological Investigations, Occasional Paper 33.
Southern Illinois University, Carbondale.
Alt, Susan M. 2012 Making Mississippian at Cahokia. In The
Oxford Handbook of North American Archaeology, edited
by T. R. Pauketat, pp. 497–508. Oxford University Press,
Oxford.
Bayliss, Alex, Johannes van der Plicht, Christopher Bronk
Ramsey, Gerry McCormac, Frances Healy, and Alasdair
Whittle 2011 Toward Generational Time-Scales: The
Quantitative Interpretation of Archaeological
Chronologies. In Gathering Time: Dating the Early
Neolithic Enclosures of Southern Britain and Ireland, Vol.
1, edited by Alasdair Whittle, Frances Healy, and Alex
Bayliss, pp. 17–59. Oxbow Books, Oxford.
Bronk Ramsey, Christopher 2009 Bayesian Analysis of
Radiocarbon Dates. Radiocarbon 51:337–360.
Bullen, Ripley P. 1951 The Enigmatic Crystal River Site.
American Antiquity 17:142–143.
Bullen, Ripley P. 1953 The Famous Crystal River Site. The
Florida Anthropologist 6:9–37.
Bullen, Ripley P. 1966 Stelae at the Crystal River Site, Florida.
American Antiquity 31(6):861–865.
Cherkinsky, Alexander, Thomas J. Pluckhahn, and Victor D.
Thompson 2014 Variation in Radiocarbon Age
Determinations from the Crystal River Archaeological
Site, Florida. Radiocarbon 56(2):1–10.
Cordell, Ann 2016 Variability in Clay Resources from the
Vicinity of Crystal River, Citrus County, Florida. Report
on file, Florida Museum of Natural History-Ceramic
Technology Laboratory, Gainesville.
Eckert, Suzanne L. 2012 Choosing Clays and Painting Pots in
the Fourteenth-Century Zuni Region. In Potters and
Communities of Practice: Glaze Paint and Polychrome
Pottery in the American Southwest, A.D. 1250 to 1700, edited
by Linda S. Cordell and Judith A. Habcht-Mauche, pp. 55–
64. Anthropological Papers of the University of Arizona No.
75. The University of Arizona Press, Tucson.
Emerson, Thomas E. 1999 The Langford Tradition and the
Process of Tribalization on the Middle Mississippian
Borders. Midcontinental Journal of Archaeology 24:3–56.
Emerson, Thomas E., and Eve Hargrave 2000 Strangers in
Paradise? Recognizing Ethnic Mortuary Diversity on the
Fringes of Cahokia. Southeastern Archaeology 19:1–23.
Emerson, Thomas E., and Dale L. McElrath 2001 Interpreting
Discontinuity and Historical Process in Midcontinental
Late Archaic and Early Woodland Societies. In The
Archaeology of Tradition: Agency and History Before and
After Columbus, edited by Timothy R. Pauketat, pp. 195–
217. University Press of Florida, Gainesville.
Garraty, Christopher 2013 Social Identity and Political
Competition in a Culturally Diverse Landscape: Decorated
Pottery from the Mescal War Site, Southeastern Arizona.
Journal of Arizona Archaeology 2(2):163–177.
Goggin, John M. 1952 Space and Time Perspective in Northern
St. Johns Archaeology, Florida. Yale University Publications
in Anthropology 47. New Haven, CT.
Goodbred, Stephen L., Eric E. Wright, and Albert C. Hine 1998
Sea-Level Change and Storm Surge Deposition in a Late
Holocene Florida Salt Marsh. Journal of Sedimentary
Research 68(2):240–252.
Gosselain, Olivier P. 2000 Materializing Identities: An African
Perspective. Journal of Archaeological Method and Theory
7:187–217.
Jones, Siân 1997 The Archaeology of Ethnicity: Constructing
Identities in the Past and Present. Routledge, New York.
Katzmarzyk, Cheryl 1998 Evidence of Stress in a Precolumbian
Population from Mound G at the Crystal River Site, Florida.
Unpublished Master’s thesis, Department of Anthropology,
University of Florida, Gainesville.
Kemp, Kassie C. 2015 Pottery Exchange and Interaction at the
Crystal River Site (8CI1), Florida. Unpublished Master’s
thesis, Department of Anthropology, University of South
Florida, Tampa.
Lave, Jean, and Etienne Wenger 1991 Situated Learning:
Legitimate Peripheral Participation. Cambridge University
Press, Cambridge.
Lyons, Patrick D., and Jeffrey J. Clark 2012 A Community of
Practice in Diaspora: The Rise and Demise of Roosevelt
Red Ware. In Potters and Communities of Practice: Glaze
Paint and Polychrome Pottery in the American Southwest,
A.D. 1250 to 1700, edited by Linda S. Cordell and Judith
A. Habcht-Mauche, pp. 19–33. Anthropological Papers of
the University of Arizona No. 75. The University of
Arizona Press, Tucson.
Marquardt, William H. 2014 Tracking the Calusa: A
Retrospective. Southeastern Archaeology 33:1–24.
McFadden, Paulette S. 2015 Late Holocene Coastal Evolution
and Human Occupation on the Northern Gulf Coast of
Florida. Unpublished Ph.D. dissertation, Department of
Anthropology, University of Florida, Gainesville.
McFadden, Paulette S. 2016 Coastal Dynamics and Pre-
Columbian Human Occupation in Horseshoe Cove on the
Northern Gulf Coast of Florida, USA. Geoarchaeology: An
International Journal 31(5):355–375.
Menz, Martin W. 2015 Like Blood from a Stone: Teasing out
Social Difference from Lithic Production Debris at
Kolomoki (9ER1). Unpublished Master’s thesis,
Department of Anthropology, University of South Florida,
Tampa.
Milanich, Jerald T. 1999 Introduction. In Famous Florida
Sites: Crystal River and Mount Royal,editedbyJeraldT.
Milanich, pp. 1–28. University Press of Florida,
Gainesville.
Minar, Jill C. 2001 Motor Skills and the Learning Process: The
Conservation of Cordage Final Twist Direction in
120 T. J. PLUCKHAHN ET AL.
Downloaded by [154.16.6.163] at 18:55 09 December 2017
Communities of Practice. Journal of Anthropological
Research 57:381–405.
Moore, Clarence Bloomfield 1903 Certain Aboriginal Mounds
of the Central Florida West-Coast. Journal of the Academy
of Natural Sciences of Philadelphia 12(2):361–438.
Moore, Clarence Bloomfield 1907 Crystal River Revisited.
Journal of the Academy of Natural Sciences of
Philadelphia, Second Series 13(3):406–425.
Moore, Clarence Bloomfield 1918 The Northwestern Florida
Coast Revisited. Journal of the Academy of Natural
Sciences of Philadelphia, Second Series 16(4):514–581.
Norman, Sean P. 2014 Modeling the Relationship between
Climate Change and Landscape Modification at the
Crystal River Site (8CI1), Florida. Unpublished Master’s
thesis, Department of Anthropology, University of South
Florida, Tampa.
Pauketat, Timothy R. 2003 Resettled Farmers and the Making
of a Mississippian Polity. American Antiquity 68:39–66.
Pauketat, Timothy R. 2007 Chiefdoms and Other
Archaeological Delusions. AltaMira, Walnut Creek, CA.
Pauketat, Timothy R., and Neal H. Lopinot 1997 Cahokian
Population Dynamics. In Cahokia: Domination and
Ideology in the Mississippian World, edited by T. R.
Pauketat and T. E. Emerson, pp. 103–123. University of
Nebraska Press, Lincoln.
Pluckhahn, Thomas J., and Victor D. Thompson 2017 Crystal
River: History and Process in the Archaeology of an Early
Village in the American Southeast. University Press of
Florida, Gainesville.
Pluckhahn, Thomas J., Alexander D. Hodson, W. J. Rink,
Victor D. Thompson, Robert R. Hendricks, Glen Doran,
Grayal Farr, Alexander Cherkinsky, and Sean P. Norman
2015a Radiocarbon and Luminescence Age
Determinations on Mounds at Crystal River and Roberts
Island, Florida, USA. Geoarchaeology: An International
Journal 30(3):238–260.
Pluckhahn, Thomas J., Victor D. Thompson, and Alexander
Cherkinsky 2015b The Temporality of Shell-bearing
Landscapes at Crystal River, Florida. Journal of
Anthropological Archaeology 37:19–36.
Roberts, Joanne 2006 Limits to Communities of Practice.
Journal of Management Studies 43:623–639.
Roddick, Andrew Paul 2009 Communities of Pottery
Production and Consumption on the Taraco Peninsula,
Bolivia 200 BC–300 AD. Unpublished Ph.D. dissertation,
Department of Anthropology, University of California,
Berkeley.
Russo, Michael, Craig Dengel, and Jeffrey Shanks 2014
Northwest Florida Woodland Mounds and Middens: The
Sacred and Not So Secular. In New Histories of Pre-
Columbian Florida, edited by Neill J. Wallis and Asa R.
Randall, pp. 121–142. University Press of Florida,
Gainesville.
Sassaman, Kenneth E. 2005 Poverty Point as Structure, Event,
Process. Journal of Archaeological Method and Theory 12
(4):335–364.
Sassaman, Kenneth E. 2012 Futurologists Look Back.
Archaeologies 8(3):250–268.
Sassaman, Kenneth E., and Wictoria Rudolphi 2001
Communities of Practice in the Early Pottery Traditions
of the American Southeast. Journal of Anthropological
Research 57:407–425.
Shennan, Stephen J. 1989 Introduction: Archaeological
Approaches to Cultural Identities. In Archaeological
Approaches to Cultural Identity, edited by S. Shennan, pp.
1–32. Routledge, London.
Smith, Hale G. 1951 Crystal River, Revisited, Revisited,
Revisited. American Antiquity 17:143–144.
Smith, Karen Y. 2009 Middle and LateWoodland Period Cultural
Transmission, Residential Mobility, and Aggregation in the
Deep South. Unpublished Ph.D. dissertation, Department of
Anthropology, University of Missouri, Columbia.
Thompson, Rachel E. 2016 Understanding Identity through
Ceramic Analysis at the Crystal River and Roberts Island
Sites. Unpublished Master’s thesis, Department of
Anthropology, University of South Florida, Tampa.
Thompson, Victor D., Thomas J. Pluckhahn, Oindrila Das,
and C. Fred T. Andrus 2015 Assessing Village Life and
Monument Construction (cal. AD 65–1070) along the
Central Gulf Coast of Florida through Stable Isotope
Geochemistry. Journal of Archaeological Science: Reports
4:111–123.
van der Leeuw, Sander E. 1984 Pottery Manufacture: Some
Complications for the Study of Trade. In Pots and Potters:
Current Approaches in Ceramic Archaeology, edited by
Prudence Rice, pp. 55–70. Institute of Archaeology,
University of California, Los Angeles.
van der Leeuw, Sander E. 1993 Giving the Potter a Choice:
Conceptual Aspects of Pottery Techniques. In
Technological Choices: Transformation in Material Culture
from the Neolithic to Modern High Tech, edited by Pierre
Lemonnier, pp. 467–553. Routledge, London.
Wallis, Neill J., and Paulette S. McFadden 2016 Recovering the
Forgotten Woodland Mound Excavations at Garden Patch
(8DI4). Southeastern Archaeology 35(3):194–212.
Wallis, Neill J., Paulette S. McFadden, and Hayley M. Singleton
2015 Radiocarbon Dating the Pace of Monument
Construction and Village Aggregation at Garden Patch: A
Ceremonial Center on the Florida Gulf Coast. Journal of
Archaeological Science: Reports 2:507–516.
Weisman, Brent R. 1995a Crystal River: A Ceremonial Mound
Center on the Florida Gulf Coast. Florida Archaeology Series
No. 8, Division of Historical Resources, Florida Department
of State, Tallahassee.
Weisman, Brent R. 1995b The Roberts Island Shell Mound
Complex and Its Archaeological Significance. Manuscript
No. 4365, Florida Master Site File, Division of Historical
Resources, Florida Department of State, Tallahassee.
Wendrich, Willeke 2012a Archaeology and Apprenticeship:
Body Knowledge, Identity, and Communities of Practice.
In Archaeology and Apprenticeship: Body Knowledge,
Identity, and Communities of Practice, edited by Willeke
Wendrich, pp. 1–19. University of Arizona Press, Tucson.
Wendrich, Willeke 2012b Recognizing Knowledge Transfer in
the Archaeological Record. In Archaeology and
Apprenticeship: Body Knowledge, Identity, and
Communities of Practice, edited by Willeke Wendrich, pp.
255–262. University of Arizona Press, Tucson.
Wenger, Etienne 1998 Communities of Practice: Learning,
Meaning, and Identity. Cambridge University Press,
New York.
Willey, Gordon R. 1949 Archeology of the Florida Gulf Coast.
Smithsonian Miscellaneous Collections 113, U.S.
Government Printing Office, Washington, D.C.
SOUTHEASTERN ARCHAEOLOGY 121
Downloaded by [154.16.6.163] at 18:55 09 December 2017
