Ritual at the Mill Cove Complex
Realms beyond the River
Copper plates, long-nosed god earpieces, and spatulate celts are not oen thought
of as the material possessions of foragers, particularly ones living at the edge of the
early Mississippian world. But in northeastern Florida, St. Johns II (AD –)
sher-hunter-gatherers acquired appreciable quantities of stone, metal, and other
mineral artifacts from far-o lands. e majority of these nonlocal items appear
to have been consumed at the community level through mortuary ritual at two
ceremonial centers: Mill Cove Complex and Mt. Royal. What is rarely mentioned,
however, is the fact that these St. Johns II societies also interred artifacts from much
earlier periods of manufacture in these same mortuary mounds. Early Mississippi
period inhabitants of Mill Cove were fully aware of the antiquity of these objects
and their associated mounds and middens. We propose herein that their acquisi-
tion, use, and burial in St. Johns II mounds helped forge a connection to a deep and
sacred past. In this chapter we explore the use of exotica and pieces of the ancient
past as a fundamental part of St. Johns mortuary ritual from the vantage point of
Kinzey’s Knoll at the Mill Cove Complex.
The Mythical Past
Evidence of past cultures is present everywhere across the landscape, and through-
out time and place societies have been guided in the present by their past (Lowen-
thal ; Gosden and Lock , ). As a product of human action, culture is
incontrovertibly historical in that it is continually negotiated and passed down
from generation to generation. Because of the ongoing process of its construction,
culture is not reproduced in its exact form, although human actions are certainly
constrained by the structural context from which they derive. In fact, it is history
that bestows the structural possibilities for human agency or any action that can
Ritual at the Mill Cove Complex: Realms beyond the River 263
conceivably run contrary to the norms, practices, and rules of society. us, all hu-
man action builds upon what came before. In this chapter, we are concerned more
with the deep past, a time that existed beyond the lifetime or memory of any one
person in a society.
Memory allows us to recall stored information and bring it forth into the pres-
ent. It works at both individual and group scales, which means that a vast array of
memories can coexist in any given community. Here, our interest centers on social
memory or “a collective notion . . . about the way things were in the past” (Van
Dyke and Alcock , ). Memories are not static or xed reections of the past;
their meanings are continually subject to negotiation, contestation, and redenition
(Lowenthal , ; Wilson , ). As the gap between the past and present
widens, the past is reconstituted using whatever evidence is available. In Florida,
material clues of primordial times would have existed in the form of mounds, mid-
dens, and durable artifacts as well as natural landscape features such as rivers, creeks,
and distinct landforms. With this in mind, a distinction can be made between ge-
nealogical history and mythical history, although the two are not necessarily mutu-
ally exclusive. In the former, the past is constructed through ties to known ancestors,
whereas in the laer deeper descent lines and “a less well-known past is evoked”
(Gosden and Lock , ). A mythical history constitutes a peoples’ notion of
what happened before the known past.
In Native American societies, history is recounted through kinship rendering
and the repeated oral telling of stories and events associated with those individuals.
e memory of the living is recreated through various physical and verbal mnemon-
ics that serve to restate the past and keep the memory alive within a society (Gosden
and Lock , ). Social memory is retrieved through a variety of means such as
song, story, dance, and art, and it is oen collectively embodied in the landscape, in
architecture, and in portable objects. But what about the ancient past, the long ago
that has faded from everyone’s memory? is is a past that is open to interpretation;
it is a mythical history that is constructed to serve the needs of those in the pres-
ent. As Lowenthal (, ) states, “the prime function of memory . . . is not to
preserve the past but to adapt it so as to enrich and manipulate the present.” In the
following we explore how St. Johns II societies of northeastern Florida may have
used mortuary ritual and “pieces of the past” to authenticate what they perceived as
tangible connections to mythical ancestors.
Mill Cove Complex
e Mill Cove Complex is the largest early St. Johns II selement in northeastern
Florida. Mapped onto rolling relict dune elds along the south bank of the St. Johns
River, Mill Cove consists of a complex of residential shell middens, special event or
K. Ashley and V. Rolland264
ritual middens, earthen causeways, and sand burial mounds. Many of the sacred fea-
tures are graed onto natural rises that allow the monuments to change shape and
height depending on one’s viewpoint on the landscape. e site’s inhabitants were
aorded ready access to daily life-sustaining resources in the form of sh, shellsh,
reptiles, and land mammals. Wild plants, nuts, and fruits were procured, but farm-
ing was not practiced. Far-reaching social relationships ensured a ow of ideas and
nonlocal materials, which meant that St. Johns II societies were never isolated from
events and developments of the early Mississippian world (Ashley , ).
e most salient features of the Mill Cove Complex are the Grant (DU) and
Shields (DU) mounds, situated approximately meters from one another
(Ashley a; unen ). Much of what we know about the two mounds
comes from the late-nineteenth-century excavations of C. B. Moore (a, b,
). Away from the two sand monuments, grid-based, shovel testing, and the
excavation of a limited number of larger investigative units have taken place over the
past years (see Ashley a, – for a review). Radiometric assays date St.
Johns II occupations at Mill Cove to ca. AD –.
e Grant and Shields mounds were erected atop the two highest points of the
site (gure .). Grant Mound was built at the edge of a -meter blu fronting the
St. Johns River, whereas Shields was situated along the top of a natural dune ridge
about meters from the river. Moore (, ) described the shape of Grant
Mound as “the usual truncated cone” and recorded its height at about eight meters.
Shields Mound was dierent: “a great platform mound entirely unlike in form any
aboriginal earthwork on the [St. Johns] river” (ibid., ). Although not as tall as
Grant Mound, at six meters in height, Shields Mound was long and narrow with a
platform summit at one end (northeast) and a shhook-like feature at the other end
(southwest). Figure . shows Moore’s diagram of Shields next to a recent LiDAR-
derived topographic map. e laer reveals the at-topped northern end, described
by Moore, but it depicts a more conical mound at the opposite end of what appears
to be a linear relict dune. At this point it is unclear whether the discrepancy is due
to landform changes since Moore’s visit or to cartographic errors on the part of
Moore’s mapper. e architecture of Shields and Grant mounds was quite dierent,
which means that they were not twin pillars bracketing the site.
Each mound was an accretionary cemetery with a long history of soil addition
and reconguration as new burials were added. Moore’s notes and stratigraphic
prole maps of Grant Mound drawn by unen () in the s reveal that
distinct episodes of burial were marked by colored sands, particularly zones that
were intentionally tinted with varying shades of red, depending on the amount of
hematite added to white or pale yellow sand. is articial coloring would have en-
tailed thorough grinding of ne-grade, nonlocal iron oxide. Local ferruginous nod-
Figure 13.1. e Mill Cove Complex showing the location of Grant Mound, Shields Mound, and Kinzey’s Knoll.
K. Ashley and V. Rolland266
ules consist of lile more than loosely concreted sand with ferric staining that would
not have produced the impressive volume of silt-sized particles of red dust observed
in mound proles. e ritual acts of processing the red sands and layering them in
mortuary mounds can be traced back to the early Woodland times in northeastern
Florida (see also White this volume). e basal portion of both mounds consisted
of deposits of pure white and black organic sands, natural soils that are readily avail-
able in the Mill Cove area.
Acts of monumentality at the Mill Cove Complex appear to have commenced
with the accumulation of a shell midden; a practice that contrasts sharply with the
premound surface preparations associated with earlier sand mounds in the region.
e building of local Woodland period Swi Creek mounds was preceded by land
clearing and the removal of topsoil followed by the deposition of a layer of charcoal-
laden white sand. Wallis (, ) interprets this as a conscious act of purica-
tion that symbolically marked “new historical beginnings.” In contrast, portions of
Grant and Shields mounds were built directly on top of preexisting St. Johns II shell
middens, and rather thick middens at that (Moore , ; unen ). is is
also the case at the Grand (DU) and Goodman (DU) mound sites (Ashley et
al. ; Jordan ; Recourt ). If St. Johns II groups were new to far north-
eastern Florida, as has been previously suggested on the basis of selement distri-
bution and ceramic evidence (Ashley , –; Ashley b, –), the
Figure 13.2. Shields Mound. Le: Florida Geographic Data Library LiDAR-derived topo-
graphic map with 2-foot contour intervals. Right: Map from Moore 1895, 10.
Ritual at the Mill Cove Complex: Realms beyond the River 267
creation of refuse deposits may have served as a way for the new arrivals to establish
a community connection to the territory.
ese underlying shell middens clearly predate the onset of mound construc-
tion but perhaps only by weeks, months, or a few years. A question arises about
the circumstances of their deposition: Is it refuse related to founding events, the
by-product of feasting, or merely trash just lying around? In light of what has been
recovered at Kinzey’s Knoll (see below), we see no reason to discount the possibility
that each mound began with commemorative feasting. We need to consider the pos-
sibility that the juxtaposition of the two mounds and shell middens was a culturally
informed choice and not simply a coincidence. In eect, they preset the landscape
for a cemetery they knew they would eventually need. Although inspirations for
mound building may have derived from elsewhere in time and/or space, the actual
social practice of constructing the mound and processing and interring the dead was
enacted locally at a specic moment in history (Wallis , –).
Mound artifacts at the Mill Cove Complex are highlighted by elaborate foreign
goods such as the two ground stone spatulate celts from Shields Mound and a set of
copper long-nosed god maskees from Grant Mound (Moore , , ). Other
nonlocal items in both mounds included mica, galena, faceted and globular quartz
crystals, ground stone celts, steatite pipes, variously shaped small copper plates, and
other copper-covered items made of stone, bone, wood, and shell. ese materials
came from far-ung areas of eastern North America; cultural centers such as Macon
Plateau, Georgia, and Cahokia, Illinois, are strongly implicated in their movement
(Ashley , –).
A curious aspect of both Grant and Shields mounds, and one that has received
lile if any aention, is the inclusion of artifacts whose period of manufacture
greatly predates St. Johns II mound construction and use. ese include Archaic
bannerstones (gure .), polished stone beads, and projectile points (e.g., Ben-
ton, Turkey Tail; see Johnson and Brookes ) as well as possible Woodland pe-
riod (Hopewell-related) ground stone plummets, quartz crystals, and steatite elbow
pipes (gure .). It also is possible that some of the many ground stone celts from
Grant Mound were manufactured before the St. Johns II period. Wilcox (, –
) notes that the Archaic projectile points from the Grant and Shields mounds
display highly symmetrical shapes with minimal evidence of retouching, perhaps
suggesting a preference for idealized biface forms in ritual use. e lithic materials
from the Mill Cove Complex are nonlocal because the site lacks an Archaic compo-
nent and the closest known chert sources are about kilometers away.
Although they are separated by kilometers, Shields and Grant mounds are
remarkably similar to Mt. Royal in terms of placement on the landscape, construc-
tion, material content, and modes of human interment. Mt. Royal, located upriver
(south) of Mill Cove in the middle section of the St. Johns River valley, is another
K. Ashley and V. Rolland268
major St. Johns II mound center with strong ties to the early Mississippian world
(Ashley c). Access to and departure from each mound was guided by walk-
ways in the form of raised earthen berms (Ashley , ). Anyone who stands
at the edge of the steep blu fronting each of these three mounds cannot help but
recognize that each provides a similar panoramic view. At the base of the blu wall
lay a broad section of the St. Johns River that separated sacred high ground on the
Figure 13.3. Archaic bannerstones from the
Shields Mound. Photographs courtesy of the
National Museum of the American Indian
and the Smithsonian Institution.
Figure 13.4. Possible Woodland period ground stone objects from the Grant Mound: a) ground stone
pendant; b) quartz pendant; c) steatite elbow pipe. Photographs courtesy of the National Museum of
the American Indian and the Smithsonian Institution.
Ritual at the Mill Cove Complex: Realms beyond the River 269
mound side from the broad shallow marshes and web of wetland islands and creeks
that spread out from the opposite bank. ese sacred spaces were specically lo-
cated and laid out with processional approaches and public viewing in mind.
Beyond the Grant and Shields mounds is a widespread yet variable distribution of
household garbage that was accentuated by several impressive shell middens (Ash-
ley a, –). One lies near the Grant Mound and another (Reeves Rise) is
located about meters north of the Shields Mound. e most intensively tested
to date, however, is Kinzey’s Knoll. is dense shell midden, positioned on a low
rise about meters northwest of the Shields Mound, measures about meters
in diameter and has a distinctly mounded core approximately meters in diam-
eter and centimeters thick. Two contiguous but staggered -meter squares were
excavated in and (Ashley a, ), and another -meter square and
a -×--meter unit were excavated in . Although analysis of the laer units is
ongoing, preliminary statements about what was recovered are provided in the fol-
Kinzey’s Knoll has yielded a striking inventory of artifacts and raw materials that
are more indicative of a mortuary mound than a shell midden. Among its varied
contents are whole and fragmented items of both local and exotic material that
seem far too precious to have been discarded in a common trash heap. To date,
no domestic context at Mill Cove or anywhere else in the region has yielded simi-
lar foreign materials, although Ocmulgee poery and occasional chert points and
akes occur in household middens throughout northeastern Florida (Ashley ,
). Elsewhere it has been argued that Kinzey’s Knoll is the material manifestation
of mortuary ritual and feasting associated with the nearby Shields Mound (Ashley
b, –; Marrinan ; Rolland ). Several lines of evidence have been
brought to bear on this claim.
First, many of the same kinds of materials were recovered from both contexts.
More than , poery fragments have been recovered from Kinzey’s Knoll. On
the basis of rim sherd data (i.e., orice size, paste, and rim and surface treatments),
a minimum of separate vessels are represented. In the – assemblage,
roughly percent of the poery was St. Johns (mostly check-stamped and plain),
and six variations of sponge spiculate paste were represented. Grit- and grit-grog-
tempered Ocmulgee Cord-Marked wares composed percent of the assemblage.
Instrumental neutron activation analysis combined with sherd rering points to
the presence of both nonlocal Ocmulgee Cord-Marked poery from south-central
Georgia and locally produced copies at Kinzey’s Knoll (Ashley , –; Rol-
land , –). In fact, rering experiments using samples from various strati-
K. Ashley and V. Rolland270
graphic levels reveal higher frequencies of Ocmulgee poery made of local clays
in the upper levels (Rolland , ). Burnished and red-lmed wares also oc-
curred in greater than expected numbers, particularly compared to other domestic
contexts across the site.
A much higher percentage of large-mouthed and small-mouthed bowls are rep-
resented at Kinzey’s Knoll compared to other areas of the site and other domestic
contexts in the region. Both St. Johns and Ocmulgee wares exhibited minimal to
no surface arition, abrasion, or heavy sooting. Rather, their nearly pristine sur-
faces suggest limited or perhaps rst-time use prior to discard. As a result, Kinzey’s
Knoll ceramics have been interpreted as a special event assemblage (Rolland ,
Kinzey’s Knoll was also marked by a high concentration of bone and shell tools
(Penders ). More than formal bone tool fragments—most of which were
still usable—have been recovered (including the excavations), many of which
are ornaments. Bone pins dominate the assemblage, and several display intricately
engraved designs that were enhanced by working ne iron-oxide powder into the
incisions (gure .). Other modied animal bone includes awls, pendants, shark
centrum beads, dolphin teeth gravers, an antler aker, a drilled turtle plastron, and
fossilized and nonfossilized shark teeth gravers and drills (Penders ). Shell ar-
tifacts consist of discoidal, tube, and barrel-shaped beads; pendants; a scraper; a
gouge; and an abrader.
Figure 13.5. Incised bone pendant,
Ritual at the Mill Cove Complex: Realms beyond the River 271
Stone and mineral items also were recovered in atypical numbers and include
nonlocal sandstone abraders, chert points and debitage, and ground stone fragments
(Bland ). Among the laer is a basalt grinder, which appears to represent a bro-
ken and reused celt fragment embedded with iron oxide powder in one rounded
end. It likely was used to crush iron-oxide nodules into powder. Diagnostic haed
bifaces consist of seven Mississippi period Pinellas, two Cahokia Side-Notched
points (gure .), one Citrus point, and one Santa Fe projectile point; the lat-
ter two date to the Archaic period. Except for one of the Cahokia points, which is
reworked, all are in nearly pristine condition. Iron-oxide or hematite nodules were
present throughout Kinzey’s Knoll, and red pigment or dust was displayed on a vari-
ety of artifacts and occasionally on discarded oyster and clam shells. Moore (a,
b, ) encountered layers and deposits of hematite-laden sands throughout
the Shields and Grant mounds, oen in association with human burials.
Fragments of sheet copper and a rolled bead further distinguish Kinzey’s Knoll
from other areas of the site. It is generally assumed that copper artifacts arrived
in northeastern Florida in nished form. While this is apparently the case for
long-nosed god maskees and other items with a broad geographical distribution,
copper-covered bone, wood, and shell artifacts and rolled beads were likely craed
locally from previously hammered pieces of sheet copper. We interpret the pres-
Figure 13.6. Cahokia Side-Notched points, Kinzey’s Knoll.
K. Ashley and V. Rolland272
ence of copper scraps along with utilized shark teeth (with evidence of bi-rotational
wear), teeth gravers, shell scrapers, bone needles, sandstone abraders, and modied
deer antler tines as the material traces of mortuary cra production.
e quality and quantity of animal bone from Kinzey’s Knoll suggest a deposit
that was distinct from the deposits of quotidian life (Marrinan ). e ethno-
graphic literature is rife with evidence of feasting in the context of ritual or religious
events in middle-range societies (see Dietler and Hayden ). In a recent faunal
study, Parsons and Marrinan () calculated vertebrate faunal density (total es-
timated biomass divided by number of cubic centimeters excavated) for nine sites
along the Georgia Bight. In their sample, Kinzey’s Knoll yielded , grams of
biomass per cubic meter, which was slightly more than twice the biomass of the
closest context, a Late Archaic shell ring along the Georgia coast. Moreover, this
calculation does not consider the dietary contribution of oysters and other shellsh
Fish and shellsh dominate the biomass, but deer remains are more prevalent
than at other known St. Johns II contexts or in local refuse deposits dating to both
earlier and later time periods. Avian remains are well represented, and all but ve
of the species of birds identied were local or migrating waterfowl. Also docu-
mented were several unusual marine or predatory mammal species such as dolphin,
bear, gray fox, and bobcat (Marrinan ). Unique to this midden, by mere pres-
ence and frequency, were unmodied vertebrae and teeth representing eight large
cartilaginous sh. ough they tend to support the idea that the faunal assemblage
is the result of ritual feasting, Parsons and Marrinan (, ) conclude that “com-
parisons with other feasting assemblages are needed to evaluate whether faunal den-
sity can be linked to ritual behavior and feasting.”
Finally, human bone was present in the shell midden at Kinzey’s Knoll, which
implicates this locus as part of the prolonged mortuary program practiced by the
residents of Mill Cove. A large piece of an innominate bone was identied dur-
ing the eld season, and a heavily modied right femur from a robust adult
and several permanent and deciduous teeth were found in . e former is a
sha fragment centimeters long from an athletic and muscular individual (Da-
vid Steadman, personal communication ); the proximal and distal ends were
removed before the bone was deposited in the midden. A narrow U-shaped, drilled
or punched-out channel was made to remove the lesser trochanter. e edges of the
channel (maximum length of . centimeters) are slightly beveled to the interior
and display four cut marks on the surface of the unsmoothed, bumpy beveled edge.
Evidence of baering is also evident, as is a lump of charcoal (. × . millimeters)
that is embedded in the exposed spongy bone at the very proximal tip. However,
the bone was not thermally altered. Sections of the diaphysis are well polished, and
at least six short cut marks are visible along the sha. None of the cut marks were
Ritual at the Mill Cove Complex: Realms beyond the River 273
deep. While the teeth and smaller remains might have been lost during handling
and transport of skeletal remains in the nal stages of mortuary ritual, the femur
is clearly a piece of worked human bone. Regardless of the specic manner and se-
quencing of the postmortem processing of bodies, this processing involved Kinzey’s
What appears to be another similar event midden is situated meters to the
north at Reeves Rise. To date, only limited testing has been done at this location in
the form of a single unit one meter square and a -centimeter shovel test that were
dug in . Excavations have yielded abundant bone and poery, a ceramic egy
adorno (canid), and a soapstone fragment. An oyster shell from Reeve’s Rise has
been radiometrically dated to the twelh century, which suggests that it postdates
the Kinzey’s Knoll shell midden (Ashley b, ).
At present, no St. Johns II burials have been recovered from nonmound contexts at
Mill Cove, although some human bone clearly ended up in the Kinzey’s Knoll shell
midden. Current evidence indicates that all deceased members of early St. Johns II
society ended a potentially lengthy mortuary process in a sand mound. Although
Moore (a, b, ) failed to provide any estimate of the total number of
burials he encountered, his excavation summaries hint at the presence of hundreds
of burials (including single skeletal elements) in each mound. Coeval mounds in
the region are much smaller and held the remains of far fewer individuals (Ashley
, –). Moore’s description of the Shields and Grant mounds indicates
that interments displayed diering degrees of completeness and articulation, which
suggests both primary and single and multiple individual secondary interments.
According to Moore (), interments of “unnatural juxtaposition” dominated at
Shields. Mixed and disarticulated skeletal remains are common in most St. Johns
Lile is currently known about St. Johns II handling of corpses prior to nal
mound interment, but deeshing, storage, and other forms of secondary manipula-
tion evidently took place. It is unclear where bodies and bundled bones were stored
aer death—in charnel houses, on scaoldings, or in the mounds themselves. e
presence of dierent burial modes could reect group/status dierentiation or
maybe the existence of a culturally prescribed time or season for mound burial.
While it seems unlikely that burial diversity reects mobility paerns, as has been
modeled for earlier Woodland mounds (Wallis , ), the large numbers of
burials in the Grant and Shields mounds may indicate that individuals living in out-
lying selements were eventually interred there sometime aer death.
Hutchinson and Aragon (, ) remind us that “many mounds [in eastern
K. Ashley and V. Rolland274
North America] contain a mixture of fairly isolated primary supine or exed buri-
als and secondary bundle burials[,] oen with a few cremations.” is is exactly
what we see in the Grant and Shields mounds. eir interpretation of such situa-
tions points to the possibility that primary and secondary burials represent dierent
phases in a multiphase mortuary process (Hutchinson and Aragon ). Some
corpses might have been placed in a sand mound only to be exhumed and reinterred
at a later date during a commemorative event. Although we currently lack complete
knowledge of the St. Johns II mortuary program, there is lile doubt that a lengthy
liminal stage was required before a recently deceased individual transitioned into a
Trying to Tie It All Together
e Shields and Grant mounds were corporate cemeteries situated in the midst of
everyday happenings. e consignment of the dead to human-made monuments
of sand has a long history along the St. Johns River that extends back to the Middle
Archaic (Sassaman , ). But mounds were far more than a pile of sand used
to store bodies. Shields and Grant mounds were storehouses of social memory and
identity. eir physicality projected visual and conceptual constructs that allowed
the living and the dead to share time and space through recurring ritual. e epi-
sodic nature of their creation allowed a myriad of relationships and identities to
be rearmed, negotiated, or even contested. e mounds were daily reminders of
community and common history, and their sacredness and physical presence fur-
ther evoked memories of the various outcomes of social interactions played out
during past mortuary events.
e same can be said of Kinzey’s Knoll, which represents an accretional special-
event midden. Its positioning on a distinct rise in the shadow of the Shields Mound
was not inconsequential; it was a culturally negotiated use of space. Two radiocar-
bon assays, separated vertically by centimeters, produced one-sigma calibrated
dates of AD – (lower midden) and AD – (upper midden), which
suggests that the shell midden accrued rapidly during the late tenth or early eleventh
century (Ashley a, ). Much of Kinzey’s Knoll consists of thick deposits of
clean, unconsolidated shells and lile soil that appear to have been piled quickly,
as is common in Late Archaic shell rings that some interpret as the “result of feast-
ing” (Russo , ). Other sections of the midden, however, exhibit loosely con-
solidated shell with more soil, hinting at less rapid buildup at times. Based on the
density distributions of shellsh, nsh, and poery remains, Rolland (, )
suggests that the sampled section of Kinzey’s Knoll accumulated as a result of at
least two episodes of more intensive use.
e details that led to the formation of Kinzey’s Knoll may never be known with
Ritual at the Mill Cove Complex: Realms beyond the River 275
certainty, but on the basis of available data it seems reasonable to conclude that
activities included graveside feasting, preparation and purication of corpses or
skeletons, and ritual acts involving the craing and use of specic mortuary items
associated with activities centered on the adjacent Shields Mound (Ashley b).
e current absence of evidence for any kind of wall, screen, or barrier suggests that
the events at Kinzey’s Knoll were spectacles that were open (visually and audibly)
to the public. is locus fostered a complex interaction between the living, the re-
cently deceased, and immemorial ancestors. Such rituals may have been a drawing
card and a source of power for Mill Cove, operating on one scale to negotiate the
community’s local and regional identities. It is likely that a similar locus existed near
the Grant Mound; the presence of the two coeval mounds in such close proximity
suggests some form of duality or social division at the Mill Cove.
A conspicuous component of both Kinzey’s Knoll and the Shields and Grant
mounds is items of a foreign nature. ese include objects of both the contempo-
rary Mississippian world and the deep past, most of which ended their use lives
in mortuary or ritual contexts. We contend that these materials (and the social
relationships they engendered) were necessary for St. Johns social reproduction;
these items were material evidence of the connection of St. Johns groups to people
and places from far away and in days of old (Ashley , ). By the eleventh
century, in the wake of or perhaps shortly before the emergence of the Mississip-
pian megacenter at Cahokia, Illinois, St. Johns II communities became entrenched
in widespread interaction and exchange networks that reached across the greater
Southeast. It has been suggested that the copper long-nosed god maskees were
“gis handed out to people who would have forever aer been aliated with Ca-
hokia” (Pauketat , ). Other items at Mill Cove from the American Boom
include a copper-covered biconical earpiece (Grant), two Cahokia points (Kinzey’s
Knoll), and perhaps some copper plates from Shields and Grant mounds. ese
and other similar items might have been gied to St. Johns groups to win them over
as important suppliers of marine shell and other Atlantic coastal materials such as
yaupon holly and bird feathers (Ashley , ; , –).
Giing engages partners in a recurring sequence of giving and receiving that has
the potential to forge prolonged social relationships over great distances; it keeps
bonds alive between individuals and groups (Mauss ; Sahlins ). Such so-
cial actions may have been part of adoption or alliance ceremonies that transformed
strangers into allies, friends, or even ctive kin. Based on the quantities of exotica at
Mill Cove, the community appears to have been quite successful at securing social re-
lations with a variety of early Mississippian communities. Once acquired, the foreign
materials might have served as specic referents to successful social networking (Wal-
lis , ). Unfortunately, with the data at hand, it is dicult to gauge whether the
appreciable quantities of nonlocal metal, stone, and other minerals in these St. Johns
K. Ashley and V. Rolland276
II mounds were the result of relatively stable, low-intensity exchange over a long span
of time (AD –) or higher-intensity interactions over a shorter interval.
What about the pieces of the past that are present at Mill Cove? ese consist
mostly of durable lithic artifacts, both chipped (projectile points) and ground stone
implements (e.g., bannerstones, elbow pipes, steatite cooking stones), that were
originally manufactured during the earlier Woodland or Archaic periods. In early
St. Johns II societies, these objects likely helped establish a cultural continuum that
brought the past to the present and provided a tangible link to the immemorial roots
of St. Johns II society (sensu Lillios ; Gosden and Lock ). ese histori-
cal referents embodied the ancestors and mythical events that formed the basis of
their origin stories and cosmologies. A periodic display of potent objects or material
symbols that communicated incontrovertible ties to the ancestral past was needed
for relationships with remote ancestors to be preserved eectively over consecutive
generations (Lillios , ).
All of the pre–St. Johns II foreign artifacts at Mill Cove conceivably could have
come from mounds in northeastern Florida. We see no reason to dismiss the no-
tion that St. Johns people dug into centuries- or even millennia-old sand mounds
to retrieve these artifacts of the mythical past. Such endeavors undoubtedly would
have brought them into contact with human skeletal remains, linking in their minds
the artifacts, the human bones, and the mound construction to ancient peoples. It
was the histories of these people that they appropriated and made their own. Actual
consanguinity was irrelevant; these were their ancestors.
By salvaging these items directly from ancient monuments, St. Johns II societ-
ies may have helped bolster assertions of an imagined genealogical history that ex-
tended back to the days of the original mound builders. Moreover, it reinforced
their claims to the land and its resources. Perhaps recovery of ancient exotica played
a part in the reawakening of local involvement in long-distance interactions net-
works during the tenth century as St. Johns II communities tried to emulate the
accomplishments of their ancestors.
Mortuary ritual, whether at Grant Mound, Shields Mound, or Kinzey’s Knoll,
drew together dispersed community members in solidarity-avowing social gather-
ings at which assertions of ancestry and identity were renewed or transformed. e
invocation of ties to ancestors is most critical during rites of passage, particularly
the liminal phase between life and death. Summoning ancestral and otherworldly
spirits through the public display of exotica derived from the both the contempo-
rary and primordial worlds was essential to these rituals. But objects manifest no
intrinsic meaning; that quality is inscribed upon them by people. Around the world,
ritual has provided a powerful context for imbuing objects and symbols with mean-
ing. ese endowed objects, in turn, enabled people to actively construct social
memory, identity, and history among the living.
Ritual at the Mill Cove Complex: Realms beyond the River 277
In fact, pieces of the past have “no ancestral identity or cosmological authentica-
tion”; these qualities are ascribed to them by the living (Lillios ; ). Mor-
tuary rituals are authored by the living, not the dead (Barre , ), and it
was the living who carried out funeral ceremonies to create and recreate life at Mill
Cove. New practices and beliefs that were enacted to engender connections to deep
time and far-ung areas of the contemporary world were likely rendered traditional
or ancient in order to validate community identity and extant sociopolitical rela-
tions. rough time, as the community’s web of social connections began to disin-
tegrate, some members of the society may have tried to use these rituals to tout new
ideologies and create more exclusive ctive genealogical ties to important ancestors
in order to legitimate social inequalities and exclusive rights to power. But such at-
tempts by elites to hijack ideology do not appear to have taken hold at Mill Cove
during early St. Johns II times (Ashley , ; , –).
Because giing among the living produced and reproduced social relationships,
gis to the ancestors or the spirit realm may have achieved a similar goal. Although
placing these symbolically charged items in the mound would have alienated them
from the living and hidden them from public view, their importance was not lost.
rough this ritualized act, the Mill Cove community gied the materials to the
spirit/ancestor world and in the process secured their placement in a sacred arena
that objectied social history, identity and memory, in eect keeping-while-giving
(Weiner ). Out of sight was not out of mind.
A nal question to consider with respect to the material remains from Kinzey’s
Knoll is this: W hy were complete projectile points, scraps of copper, and modied and
unmodied fragments of human bone discarded there? Why were these hard-to-ob-
tain materials not recycled or placed in the nearby burial mound? Perhaps social con-
vention dictated that such symbolically charged materials could not leave designated
areas of sacred space and thus had to be ritually discarded at designated times during
or at the conclusion of the event. Maybe these objects were conned to a liminal state
in which they could not associate with the living or the ancestors. Instead of keep-
ing the objects, it seems that what was most important was keeping active the shared
acts and social links of obligation aained through acquiring these objects via giing,
exchange, or questing. Being “St. Johns” at this time may have rested partly on a com-
munity’s ability to sustain linkages that spanned long distances and extended deep
in time, and success may have been materialized through the acquisition of foreign
objects. Sand mounds and the ancestral bones and sacred objects they held referenced
not only the present and past but also future relations that communities aspired to sus-
tain or secure through far-reaching social networks (omas ; Wallis , ).
e memories created through public ritual and other commemorative performances
were as much, if not more, an invention of the social realities of early Mississippian
times as they were an accurate representation of the past.
K. Ashley and V. Rolland278
e broken bone and shell artifacts and copious amounts of fragmented poery
at Kinzey’s Knoll also are interesting, particularly because the broken objects were
still usable. Is this also merely ritual trash that was deposited there because it was
the socially required thing to do? Or was there more to their fragmented condition?
It is worth noting that while many of the sherds from Kinzey’s Knoll are quite large
and number more than ,, not one vessel has been reconstructed (wholly or
partly) from lip to base, despite extensive aempts at cross-mending. e concept
of enchainment oers a compelling possible explanation for why, in certain con-
texts, special artifacts occur in broken or unmendable conditions (Chapman ,
). Enchainment links people, places, and things through the deliberate breakage
of objects and the subsequent distribution of their parts among various households
across a potentially broad landscape. In mortuary contexts, fragmentation and en-
chainment act to cement an ideological link among the living, their ancestors, and
the not-yet born (Ashmore ; Casey ; Chapman ; Strang , Ta-
çon ). e fragments become mnemonic or metaphorical references to people
(living and dead), places, and events. It is another case of keeping-while-giving.
During culturally meaningful gatherings, Mill Cove residents and other partici-
pants may have been allowed to contribute food and other cra items. ose used
during these events, or merely those belonging to aendees, may have been broken
or fragmented and a portion deposited in the shell midden at Kinzey’s Knoll, with
the remainder retained by the craer and/or family members as a symbolic token of
a shared community experience (Chapman , ; Strang , ). Might the
basal sections or “hearts” of “killed” vessels deposited in the many mounds along
the St. Johns serve the same function, enchaining the individual poer (family or
other social group) to the ancestral past? Kinzey’s Knoll cannot be simply dismissed
as trash or an accumulation of rubbish associated with a particular gathering. Con-
tinued and expanded reing aempts in conjunction with additional excavations
at Kinzey’s Knoll are needed to eliminate sampling bias as the cause of incomplete
Five centuries before the invasion from Europe, the Mill Cove Complex was one of
the premier mound centers in Florida. e St. Johns II communities at Mill Cove
and those to the south at Mt. Royal were in the vanguard of incipient and early Mis-
sissippian interactions during the late tenth through twelh centuries. ey were
able to establish social relations with distant peoples and bring exotic raw materials
and nished products to northeastern Florida. In this chapter we argue that at Mill
Cove these items assumed mortuary and ritual signicance, eventually terminating
their use life in the Shields or Grant mound or in special-event middens such as
Ritual at the Mill Cove Complex: Realms beyond the River 279
Kinzey’s Knoll. ese exotic materials were not used for utilitarian purposes. Not
only did Mill Cove residents acquire objects from distant lands, they also secured
items from the deep past. rough ritual, these items created a community connec-
tion to ancestors and the cosmos.
Excavations at Kinzey’s Knoll have produced a tremendous number of ceramic
vessel fragments; a diverse range of modied bone, shell, and lithic artifacts; pieces
of copper; and an incredibly high density of faunal material. e amount of animal
bone per cubic meter of midden matrix is simply staggering. ese material residues
spotlight ritual behavior during the eleventh century that involved preparation and
consumption of food, burial preparation, and craing and discard of highly special-
ized materials. At sacred locations such as Kinzey’s Knoll, memories of the recent
past and deeper histories were authenticated and made public. Sand monuments
clearly were not the only arenas of ceremony at the Mill Cove Complex.
We would like to thank Kinzey and John Reeves as well as the entire Reeves family
for allowing us to work on their property. eir continued support of our research is
greatly appreciated. We thank Michael Boyles for his help with gures . and .,
Buzz unen for gure ., and Ray Eslinger for the LiDAR images. Figures .
and . were used with the permission of the National Museum of the American
Indian at the Smithsonian Institution. anks also to Neill and Asa for puing to-
gether the symposium and marshaling the volume through the publication process.
Ashley, Keith H.
On the Periphery of the Early Mississippian World: Looking Within and Beyond Northeast-
ern Florida. Southeastern Archaeology : –.
Interaction, Population Movement, and Political Economy: e Changing Social Landscape
of Northeastern Florida (A.D. –). PhD dissertation, Department of Anthropology,
University of Florida, Gainesville.
a Introducing Shields Mound (DU) and the Mill Cove Complex. Florida Anthropologist :
b Toward an Interpretation of at the Shields Mound (DU) and Mill Cove Complex. Florida
Anthropologist : –.
c An Archaeological Overview of Mt. Royal. Florida Anthropologist : –.
Early St. Johns II Interaction, Exchange, and Politics: A View from Northeastern Florida. In
Late Prehistoric Florida: Archaeolog y at the Edge of the Mississippian World , edited by Keith
Ashley and Nancy M. White, –. University Press of Florida, Gainesville.
Ashley, Keith, Vicki Rolland, and Rochelle Marrinan
A Grand Site: Testing of the Grand Shell Ring. Report on le, Division of Historical Re-
K. Ashley and V. Rolland280
Visions of the Cosmos: Ceremonial Landscapes and Civic Plans. In Handbook of Landscape
Archaeology, edited by Bruno David and Julian omas, –. Le Coast Press, Walnut
Barre, John C.
e Monumentality of Death: e Character of Early Bronze Age Mortuary Mounds in
Southern Britain. World Archaeology : –.
Bland, Myles C. P.
Moore to the Point. Paper presented at the th Annual Southeastern Archaeological Confer-
ence, Chaanooga, TN.
Casey, Edward S.
Place in Landscape Archaeology: A Western Philosophical Prelude. In Handbook of Land-
scape Archaeology, edited by Bruno David and Julian omas, –. Le Coast Press, Walnut
Object Fragmentation and Past Landscapes. In Handbook of Landscape Archaeology, edited by
Bruno David and Julian omas, –. Le Coast Press, Walnut Creek, CA.
Dietler, Michael, and Brian Hayden (editors)
Feasts: Archaeological and Ethnographic Perspectives on Food, Politics, and Power. Smithsonian
Institution Press, Washington, DC.
Gosden, Chris, and Gary Lock
Prehistoric Histories. Worl d Archaeology : –.
Gosden, Chris, and Yvonne Marshall
Cultural Biography of Objects. World Archaeology : –.
Hutchinson, Dale L., and Lorraine V. Aragon
Collective Burials and Community Memories: Interpreting the Placement of the Dead in the
Southeastern and Mid-Atlantic United States with Reference to Ethnographic Cases from
Indonesia. In e Place and Space of Death, edited by Helaine Silverman and David Small,
–. Archeological Papers of the American Anthropological Association vol. . American
Archaeological Association, Arlington, VA.
Johnson, Jay K., and Samuel O. Brookes
Benton Points, Turkey Tails, and Cache Blades: Middle Archaic Exchange in the Midsouth.
Southeastern Archaeology : –.
Jordan, Douglas F.
e Goodman Mound. In Papers on the Jungerman and Goodman Sites, Florida, edited by D. F.
Jordan, E. S. Wing, and A. Bullen, –. Contributions of the Florida State Museum Social
Sciences no. . University Press of Florida, Gainesville.
Lillios, Katina T.
Objects of Memory: e Ethnography and Archaeology of Heirlooms. Journal of Archaeologi-
cal Method and eory : –.
e Past Is a Foreign Country. Cambridge University Press, Cambridge, UK.
Early Mississippian Faunal Remains from the Shields Mound (DU). Florida Anthropolo-
gist : –.
Ritual at the Mill Cove Complex: Realms beyond the River 281
e Gi: Forms and Functions of Exchange in Archaic Societies. W. W. Norton, New York.
Moore, Clarence B.
a Certain Sand Mounds of the St. Johns River, Florida. Part I. Journal of the Academy of Natural
Sciences of Philadelphia : –.
b Certain Sand Mounds of the St. Johns River, Flor ida. Part II. Journal of the Academy of Natural
Sciences of Philadelphia : –.
Certain Sand Mounds of Duval County, Florida. Journal of the Academy of Natural Sciences of
Philadelphia : –.
Parsons, Alexandra, and Rochelle Marrinan
An Assessment of Coastal Faunal Data. In Life Among the Tides: Recent Archaeology on the
Georgia Bight, edited by Victor ompson and David Hurst omas, –. Anthropologi-
cal Papers of the American Museum of Natural History no. . American Museum of Natural
History, Washington, DC.
Pauketat, Timothy R .
Cahokia: Ancient America’s Great City of the Mississippi. Penguin Books, New York.
Penders, omas E.
Bone, Antler, Dentary, and Shell Artifacts from the Shields Site (DU). e Florida Anthro-
pologist : –.
Final Notes on the Goodman Mound. Florida Anthropologist : –.
Measuring Tradition and Variation: A St. Johns II Poery Assemblage from the Shields Site
(DU). MS thesis, Department of Anthropology, Florida State University, Tallahassee.
An Investigation of St. Johns and Ocmulgee Series Poery from the Shields Site (DU).
Florida Anthropologist : –.
Measuring Shell Rings for Social Inequality. In Signs of Power: e Rise of Cultural Complex-
ity in the Southeast, edited by Jon L. Gibson and Philip J. Carr, –. University of Alabama
Stone Age Economics. Aldine de Gruyter, New York.
Sassaman, Kenneth E.
e Eastern Archaic, Historicized. AltaMira Press, Lanham, MD.
Uncommon Ground: Landscapes as Social Geography. In Handbook of Landscape Archaeol-
ogy, edited by Bruno David and Julian omas, –. Le Coast Press, Walnut Creek, CA.
Marks of Possession: e Archaeology of Territory and Cross-Cultural Encounter in Austra-
lia and South Africa. In Handbook of Landscape Archaeology, edited by Bruno David and Julian
omas, –. Le Coast Press, Walnut Creek, CA.
e Politics of Vision and the Archaeologies of Landscape. In Landscape: Politics and Perspec-
tives, edited by Barbara Bender, –. Berg, Providence, RI.
unen, Robert L.
Grant Mound: Past and Present. Florida Anthropologist : –.
K. Ashley and V. Rolland282
Van Dyke, Ruth M., and Susan E. Alcock
Archaeologies of Memory: An Introduction. In Archaeologies of Memory, edited by Ruth M.
Van Dyke and Susan E. Alcock, –. Blackwell, Oxford.
Wallis, Neill J.
Networks of History and Memory: Creating a Nexus of Social Identities in Woodland Per iod
Mounds on the Lower St. Johns River, Florida. Journal of Social Archaeology : –.
Weiner, Annee B.
Inalienable Possessions: e Paradox of Keeping-While-Giving. University of California Press,
Wilcox, Jennifer R.
An Assessment of Archaic Haed Biface Use in Mississippian Burial Contexts. MA thesis,
Department of Anthropology, State University of New York, Albany.
Wilson, Gregory D.
Community, Identity, and Social Memory at Moundville. American Antiquity : –.