Front Cover
 Table of Contents
 Front Matter
 Colour plates
 Back Cover

Digitization of this item is currently in progress.
Caribbean Quarterly
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00099208/00094
 Material Information
Title: Caribbean Quarterly
Physical Description: Serial
Language: English
Creator: University of the West Indies
Publisher: Extra Mural Dept. of the University College of the West Indies
Place of Publication: Mona, Jamaica
Genre: serial   ( sobekcm )
 Record Information
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 843029
sobekcm - UF00099208_00094
System ID: UF00099208:00094

Table of Contents
    Front Cover
        Front Cover
    Table of Contents
        Page i
    Front Matter
        Page ii
    Colour plates
        Page iii
        Page iv
        Page v
        Page vi
        Page vii
        Page viii
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
        Page 166
        Page 167
        Page 168
        Page 169
        Page 170
        Page 171
        Page 172
        Page 173
        Page 174
        Page 175
        Page 176
        Page 177
        Page 178
        Page 179
        Page 180
        Page 181
        Page 182
        Page 183
        Page 184
        Page 185
        Page 186
        Page 187
        Page 188
        Page 189
        Page 190
        Page 191
        Page 192
        Page 193
        Page 194
    Back Cover
        Back Cover
Full Text


q'- .U

Guest Editors: Sabrina R. Rampersad and James C. Robertson.

ISSN 0008-6495

SCaribbean Quarterly
Vol. 55, No. 2
June, 2009

. .... .... .... i....i
T..., Ir
U~~i %.V i l ,,:!
fj 1.


.4 W.-

41: 3



.. .. . .




Professor, the Hon. R.M. Nettleford, O.M. Vice Chancellor E, Editor
Sir Roy Augier, Professor Emeritus, History, Mona.
Professor H. Beckles, Pro Vice Chancellor and Principal, UWI, Cave Hil.
Professor B. Chevannes, Research Fellow, Mona School of Business, UWI
Professor Wayne Hunte, PVC Graduate Studies and Research, UWI, St.
Professor B.Lalla, Faculty of Arts and Education, UWI, St.Augustine
Mr. J. Periera, Vice Principal, UWI, Mona
Professor Clement Sankat, PVC, Principal, UWI, St. Augustine
Professor Gordon Shirley, PVC and Principal, UWI, Mona.
Professor H Simmons-McDonald, PVC, Open Campus
Mrs. Linda Speth, General Manager, UWI Press
Dr. B. Tewarie, PVC, Office of Planning and Development, UWI,
Professor Alvin Wint, PVC, Board for Undergraduate Studies, UWI, Mona
Dr. V.Salter, CSI, Managing Editor,

Manuscripts CQ is a review journal. Articles and book reviews of relevance to
the Caribbean will be gratefully received. Authors should refer to the guidelines
on the web page or in any volume of CQ

Articles submitted are not returned.

Exchanges: Exchanges are conducted by the Gifts and Exchanges Section,
Library, University of the West Indies, Mona, Kingston 7, Jamaica

Back Issues and Microfilm: Information for back volumes supplied on request.

Caribbean Quarterly is available on microfilm from Xerox University
Microfilms and in book form from Kraus-Thompson Reprint Ltd.

Recent volumes are available online from EBSCO

Abstract and Index : 1949-2006 Author Keyword and Subject Index available
as a hard copy and also on the website.

The journal is abstracted by AES and indexed by HAPI

All correspondence and contributions should be addressed to:
The Editor,
Caribbean Quarterly,
Cultural Studies Initiative, OCVC-E
University of the West Indies,
PO Box 130, Mona, Kingston 7,
Tel. and fax No. 876-970-3261, Email: cgquwimona.edu.jm


'- .*

. .* .. A.:



DeWaal: p.16

Plate 1: Carved limestone zemi (c.8.5 cm base length) from Site du Phare

Plate 2: Carved Strombus gigas spatula (c.9 cm long) and carved and
incised Cittarium pica fishhook from Est de Mouton de Bas

Golding-Frankson: p .58

Plate 3: Two of the columella openings with inserted ovoid stone (Author's photograph)

White: p.76

a C D E F G





Plate 4: KMK2 grid placed at periphery of mound. numbers in unit are the final
excavated depth.

L* I~. ., -

~ .J .~n ____


__ _

White p.85

Non-Ceramic Artifacts




Metal Lithics Glass Organics Other

Plate 5: Comparison of non-ceramic artifacts from KMK1 and KMK2

Body Base Handle UID

Plate 6: Distribution of morphological ceramic parts recovered from KMK1 and KMK2



Morphological Parts of Ceramics




Count 600



-- IKMK2

Robertson: Plate 7, p.120: Plate 8, p. 121

Plate 7: The Assembly of Jamaica, Spanish Town Square (Photograph, Sabrina

Plate 8: Rodney Monument and its associated buildings (Photograph, Birte Timm).

Francis -Lindsay: Plate 9, p. 156, Plate 10, p. 160

I Hokkaido





Plate 9: Map of Japan. Shirakawa-go and Gokayama are located in the Chubu region
(pictured in brown). Source: About cor-Japan-Travel, www.gojapan.comics/japanmaps.

Plate 10: Local residents ensure the sustainability of the site's pristine environment by
setting "guidelines" for visitors (Photograph: K Suemori).

I -<


Caribbean Quarterly, Volume 55, No. 2, June, 2009, is a special issue
entitled : Caribbean Archaeology and Material Culture. It is appropriate that
this area of Caribbean study is being addressed for the first time by Caribbean
Quarterly as the Departments of History at Cave Hill, St. Augustine and Mona
have units for the teaching of archaeology, and there is a shortage of good material
for students to use. The issue contains seven articles and as many book reviews all
pertaining to the topic. The articles were presented originally at the
Archaeological Society of Jamaica Symposium held at Mona in 2008. The articles
are diverse in scope and explore hitherto un-researched areas from'shellsmithing'
techniques used by the Tiano to make artefacts and adornments, and the yabba
trade in Jamaica, to linkages between Maroons and Tiano peoples in Suriname.
There are also articles on the built environment and one offering possibilities for
developing Falmouth, Jamaica into a heritage tourism site modelled off the
approach used very successfully by the Japanese. This is timely as there are plans
afoot to build a massive pier to dock the largest cruise ship in the world at
Falmouth. Many environmentalists will no doubt see this as a threat to our
valuable heritage for the sake of providing potable water and sewage disposal for
the 6000 day visitors who will not be staying in our hotels. archaeology could well
point tourism in another direction. There are vast areas where supervised 'digs'
could be arranged for the discerning traveller as well as for the peoples of the
region. So far, very little is known about the original inhabitants or the probable
interactions that took place between these survivors in remote areas in the
Caribbean and the arrivants also forced to leave their homelands. Caribbean
Quarterly welcomes the initiative and editorial guidance of Sabrina Rampersad,
head of archaeology section, Mona campus together with James Robertson, from
the Department of History and Archaeology, Mona.

Caribbean archaeology and material culture

This special issue of Caribbean Quarterly is the first to be devoted to
archaeology. It is intended to demonstrate both the extensive opportunities for
archaeological work across the Caribbean as well as some of the variety of
approaches available to tackle questions concerning the region's archaeology. We
hope that archaeologists--or would-be archaeologists-will be encouraged by the
interesting work being undertaken by their colleagues in neighboring territories,
while non-archaeologists will gain some idea why the region's archaeologists
extol the value of their discipline's insights as contributions to wider debates.
These papers originated from a one-day symposium hosted by the
Archaeological Society of Jamaica (ASJ) in April 2008, which encouraged
submissions dealing with topics from the wider Caribbean. The contributions to
this issue were recruited by the editors not only for their wide range of scholarly
interests, but also to show the ASJ's commitment to embracing and promoting
varied archaeological pursuits. The combined offerings present prehistoric and
historic contexts where diverse archaeological approaches are applied. These
include: field survey and excavation (De Waal and Rampersad), pre-Columbian
manufacturing techniques and experimental archaeology (Frankson),
anthropological archaeology and ethnography (White), scientific analysis of
material culture (Hauser), the use of archival sources to reconstruct historic space
(Robertson), historical and feminist analyses of material culture (Josephs), and
comparative heritage management and tourism (Francis-Lindsay). The editors
believe that this collection's strength lies in its thematic diversity, which should
allow it to serve both as an educational tool, and to offer models for current and
future developments in Caribbean archaeology. The individual papers not only
showcase multiple approaches within the discipline, but together they demonstrate
the wider range of opportunities for cross-disciplinary insights while exploring
"the quirky borderlands between overlapping disciplines and data sets" (Mayne,
2008, 104).
The 2008 conference was the sixth Symposium that the ASJ has organized.
Like its predecessors, it attracted scholars from within Jamaica and the Caribbean,
and beyond, this time as far away as the United States and Japan. The role of the
ASJ in assembling such a broad representation of scholars and archaeological
topics is characteristic of the Society's tradition of encouraging comparative and
cross-disciplinary insights. The days when a leading British archaeologist would

write that "Jamaican archaeology is still very little known, probably for lack of
properly controlled excavations" are, fortunately, long past (Crawford, 1955, 37;
also, Robertson, 2007). Modern Jamaica is adopting pan-Caribbean and global
outlooks with regard to its archaeology and heritage management schemes. These
approaches are reinforced by the activities of local organizations like the ASJ,
which began in 1965 as the 'Archaeological Club of Jamaica,' and was
reorganized as the Archaeological Society of Jamaica in 1970. Interdisciplinarity
has always been central in this small society, all the more so when both the Club
and the Society owed much of their early energy not to an archaeologist, but to a
professional geologist, Dr. James Lee (1924-2006), who was himself a keen
avocational archaeologist. From the start, the Society encouraged a range of local
field projects and liaisons with international scholars (Lee, 1978). Dr. Lee not only
led the Society himself for its first seventeen years, but made valuable
contributions to the study of Jamaica's prehistory, including a thorough survey of
Taino sites across the island which present-day scholars continue to rely upon as a
basic starting point. Its seems only fitting that as editors and, indeed, as members
of the ASJ, we should offer our respects to the Society's founding members as the
ASJ turns toward the challenges of twenty-first century archaeology.

This symposium offered a further milestone in the development of
Caribbean archaeological research, as it successfully brought together the
archaeology faculty from the three campuses of the University of the West Indies
(St. Augustine, Trinidad and Tobago, Cave Hill, Barbados, and Mona, Jamaica).
We have finally arrived at an age in which all three of the UWI campuses now
offer archaeology at the undergraduate level. Archaeology was initially offered at
the St. Augustine campus between 1980 and 1988 and, after a 13-year hiatus,
classes were re-established there in 2001. The subject has been taught at the Mona
campus since 1987, while the Cave Hill campus in Barbados has come on board
most recently, with a full teaching program implemented in 2007/8 by its newly
hired archaeology lecturer. We hope that all three university faculty members, in
addition to the graduates of their programs, will continue to initiate
locally-directed research programs, which should in turn help to transform the
longer-term development of West Indian archaeology. In an issue of the
University of the West Indies's flagship cultural journal we are particularly happy
to have representation from not only all three campuses but also from Jamaica's
Northern Caribbean University and the Caribbean School of Architecture at the
University of Technology, Jamaica. We hope that this will be the first of many
such issues to present cross-regional and inter-disciplinary collaborations on the
region's remarkable archaeological resources.

As editors we are pleased to have secured all of the contributions in this issue
and are impressed that our contributors managed to get their material to us in a
timely fashion, despite some daunting obstacles. We are also grateful for those
individuals who took the time to produce the book reviews of several valuable
studies in Caribbean archaeology, and to Alison West-Martin, who not only
displayed her art (inspired by Taino archaeology) at the Symposium, but has
kindly allowed two items to be reproduced here (Cover and Frontispiece).

Crawford, O. G. S. (1955). Said and done: The autobiography of an archaeologist.
London: Widenfeld.
Lee, J. W. (1978). Current archaeological activity in Jamaica. Archaeology Jamaica,
78,(4), 1-4.
Mayne, A. (2008). On the edges of history: Reflections on Historical archaeology.
American Historical Review, 113, (1), 93-118.
Robertson, J. C. (2007). On 'Mummies' and 'Caribs': O.G.S. Crawford's visit to Jamaica
in 1907. Archaeology Jamaica, 20, n.s., 8-10.

Guest Editors

Small Islands, Large Settlements: Archaeology
at Les miles de la Petite Terre, Guadeloupe,


Between 1998 and 2000, three archaeological fieldwork campaigns were
carried out at the islands of Petite Terre, F.W.I., by teams of archaeologists from
Leiden, the Netherlands. The investigations, directed by the author, then working
at Leiden University, were part of a study that was designed to investigate
pre-Columbian social organisation and interaction through the study of site
patterns (De Waal, 2006). For this study, a micro-region was selected, consisting
of the Pointe des Chateaux peninsula of Guadeloupe and the islands of La
Desirade and Petite Terre (Figure 1). This essay focuses on the pre-Columbian
occupation of the small islands of Petite Terre.
Les miles de la Petite Terre, as the islands are called officially, belong to the
Lesser Antilles. They will be referred to as Petite Terre, as is common practice on
Guadeloupe. Petite Terre is part of the French department of Guadeloupe and
administratively it belongs to the municipality of La Desirade. Petite Terre is
situated at approximately 12 km south of La Desirade and at 7.5 km south-east of
Pointe des Chateaux.
Petite Terre consists of two small calcareous islands: Terre de Haut and
Terre de Bas (Figure 2). Terre de Bas measures 2.5 km by 600 m, while Terre de
Haut measures only 1.1 km by 200-300 m. The islands of Petite Terre are flat
west-east oriented islands that originally consisted of one elevated coral plateau.
They show a general inclination in a west/northwestern direction, which is the
result of tectonic processes (DIREN, Guadeloupe, 1994, 6). A 150 m wide
channel, which is 7 m deep at most and is enclosed on the eastern side by an
impressive reef barrier, separates the islands.

Figure 1. Map of the Eastern Caribbean and location of the research area (detail).

0 10km

t .g. .,..ur- .

1 1 BEfnni

"' tedar.



Les Saintes

The islands are almost completely enclosed by coral reefs (DIREN
Guadeloupe, 1994, 6). Abrasion of these reefs by sea action has created sand
beaches that dominate the low northern and western parts of Terre de Bas. These
sandy areas are easily influenced by wind and sea action. They are characterized
by dynamic dune formations that border a depression that nowadays contains
salinas (Conservatoire du Littoral, 1997, 8; DIREN Guadeloupe, 1994, 6). At high
seas, for example during hurricanes, the easternmost of these salinas is in open
connection to the sea. Therefore, it is always filled with water, whereas the two
others are periodically dry (Rousteau, 1995, 8). The more elevated south and east
coasts of Terre de Bas are rocky with limestone outcrops, as is most of Terre de

Figure 2: The islands of Petite Terre, with the location of the salinas on Terre de Bas.


Coastal dynamics, mainly caused by wave action, have drastically altered
the coasts of Petite Terre since pre-Columbian times. Coastal areas with
reef-passages or which lack reef protection have been subjected to heavy coastal
erosion. Tropical storms and hurricanes greatly accelerate erosion processes.
Areas with salinas and sand beaches are the least stable and most vulnerable to
erosion. The northern and western coasts of Terre de Bas are particularly unstable
with large parts of the beach and the dunes eroding rapidly (Figure 3). This is
evidenced by steep, almost vertically eroded, slopes of the dunes (personal
observation 1997-2000). Although a lot of the sandy coast disappears, an
important part of the sand is displaced inland as well. The dunes on Petite Terre,
which are relatively young and which appear to be shifting rapidly, are
characterized by pioneer vegetation. The salinas are gradually being filled up with
sand, a process that is clearly visible at the easternmost of the three principal
salinas of Terre de Bas (Rousteau, 1995, 8).

Figure 3: Surveyed areas and possibly eroded coastal zones on Petit Terre

*NO Su-wdw.
:::::: Posaibly n1ded o altazonlr O oom

Possibilities of obtaining fresh water on Petite Terre appear limited, as most
fresh water sources are not currently visible from the surface. Many of these are
fresh water lenses, or aquifers, situated between chalky layers in the subsurface, or
floating on saline groundwater as a result of their lower density. This
phenomenon, fed by rainfall, has also been reported for other flat limestone
islands, for example Barbuda (Watters, 1980, 64). According to Antczak (1998,
109) Los Aves fishermen still excavate pits in the dunes to obtain brackish but
drinkable water. Some of these pits can provide water throughout the year. The
presence of a well at Petite Terre demonstrates recent use of such lenses (see
Figure 2). In pre-Columbian times, the lenses could be exploited through digging
as well. In addition, on the limestone plateaus of the central parts of the islands of
Petite Terre waterholes containing rainwater have eroded (Rousteau, 1995, 9).

At Petite Terre today, soils are poorly developed, but extensive
slash-and-bur horticulture, common to agricultural Amerindian societies, is
possible on both islands, except of course in the sandy beach areas. The islands are
covered by a coastal vegetation (mainly consisting of sea grape) and very dense
dry, thorny 'inland' vegetation, mainly consisting of acacia and manchineel trees
(Barr6 et al., 1997, 7, 8 and map 2; Conservatoire du Littoral, 1997, 15).
The islands are presently characterized by an abundant and diverse
terrestrial fauna, including iguanas, turtles, land crabs and rats, the latter being
accidentally introduced by colonial occupants of the region (DIREN Guadeloupe,
1994, 15). The central channel has until recently been very rich in sea urchins,
Strombus gigas (queen conch) and a wide variety of coral species. Close to the

bordering reefs parrotfish and barracuda can be caught in abundance. Sharks are
usually found in the deeper areas south of the islands, and whales may pass the
islands between March and May (DIREN Guadeloupe, 1994, 17-20). The areas
around the Terre de Bas salinas are reputed to attract many sedentary birds and
migrating birds (Barr& et al., 1997, 19, 24).

Colonial and Recent History
The islands of Petite Terre are uninhabited today, but in the eighteenth
century they were used intensively for growing cotton. From 1826 onward, the
islands were the property of a family from La D6sirade. In 1858, twenty-eight
persons lived on the islands and by the 1940s, there were seven families. Despite
the apparent lack of fresh water, the Petite Terre gardens produced sufficiently
large yields to sell part of the products on La Desirade. In addition, the Petite Terre
inhabitants preserved their fish with salt from the local salinas. They kept sheep,
goats and pigs and caught numerous turtles that came on land to lay their eggs.
Sufficient amounts of fresh water were collected in the cistern near the lighthouse.
The last inhabitants of the islands were the lighthouse-keeper and his wife. They
left the islands in 1974.
Since 1994, the islands of Petite Terre have been state property and a nature
reserve. The coastal zone of approximately 80 m wide is managed by the Office
National des F6rets (ONF). The remainder of the islands has been controlled by
the Conservatoire de I 'Espace Littoral et des Rivages Lacustres (CELRL) and by
the Direction Departementale de l'Equipement (DDE); (Barr6 et al., 1997, 6;
Conservatoire du Littoral, 1997, 6).
Earlier Research
The first archaeological fieldwork conducted at Petite Terre was probably
carried out in 1965 or 1968. Unfortunately, there is no documentation of this
fieldwork and it is not known who actually carried out these excavations and
which site was investigated. A large amount of artefacts in the depot of the
archaeological museum on Guadeloupe testify to this unidentifiable research
The first fieldwork that was briefly documented was carried out by
Desmond Nicholson (1975). He reported a short visit to the Site du Phare, which is
on Terre de Bas. During this visit he excavated a small test unit and collected pottery,
which he labelled 'Carib' and which is presently stored at Yale University.
In 1984 the islands were visited by Pierre Bodu, a French archaeologist who
fulfilled his military service by working for the archaeological museum on

Guadeloupe. Bodu carried out site surveys on five sites on Petite Terre and
documented these (Bodu, 1985).

1998-2000 Field Investigations

At the start of the writer's 1998-2000 investigations, it had been
hypothesised that the study of settlement patterns in the micro-region consisting of
Pointe des Chateaux, La D6sirade and Petite Terre could reveal information on
pre-Columbian interaction in this area. In order to obtain a reliable archaeological
site inventory needed to provide representative information on site patterns, it was
thought that systematic and intensive surface surveys should be carried out. The
focus of the surveys was on the identification of archaeological sites (De Waal,
2006, 18). Petite Terre needed to be investigated by the Leiden teams as no
systematic or intensive surveys had been carried out on these islands before 1998.
Fieldwalking, the method of surface survey used, covered parallel transects,
separated by 10 m intervals at Terre de Haut and 20 m intervals at Terre de Bas.
Transects were 1 m wide since this is the maximum area that can be overseen at a
moderate walking speed. The effectiveness of the transect intervals was tested by a
so-called 'zigzag' survey technique, randomly checking the interval surfaces in
order to see if concentrations of archaeological material had been missed.
Transects were compass-oriented north-south covering the coastal as well as the
'inland' areas. Transect surfaces were cleared, if necessary, with machetes in order
to optimise observation. For all discovered concentrations of archaeological
material, surface collections of diagnostic artefacts were made in the field.
Although surface samples do involve selection biases and loss of provenience
information, they may be useful for attaining rough and general site
characterizations. Off-site material and surface concentrations of archaeological
material were mapped on 1:5000 aerial photographs (De Waal, 1998, 1999a).
Although the Pointe des Chateaux peninsula and La D6sirade could be
successfully surveyed using the technique outlined above, the survey of Petite
Terre was greatly hindered by the impenetrable vegetation. The dense acacia
forests and manchineel trees, which easily cause serious burning scars after
contact with bare skin and eyes, made it very difficult to cross the terrain.
Although Terre de Haut was completely surveyed at 10 m intervals during a
two-day visit by 10 archaeologists in 1998, the vegetation on Terre de Bas made it
impossible to completely survey this area intensively and systematically in the 10
days that were available to our six-person survey team in 1999.2 It was therefore
decided to focus on the sites that had been reported earlier at Terre de Bas by
Nicholson (1975) and Bodu (1985), but since surroundings and access routes to

the individual sites were intensively surveyed during the recent field
investigations, a selective but important part of the island could still be studied by
the present team (Figure 3).

Sub-surface testing was conducted at sites that we thought were deeply
stratified and not just limited to the surface (0-5 cm), and in particular those sites
which were expected to be multi-component.
The excavation of 1 square metre test units, or a series of them, was aimed at
the collection of information on geological and archaeological stratigraphy, on site
formation and deformation processes and on the chronological and cultural
context. Test unit locations were randomly chosen in areas with relatively dense
archaeological deposits. The locations, all with a north-south orientation, were
measured by GPS (Global Positioning System) and infrared theodolite. Test units
were excavated in 10 cm arbitrary levels, while taking geological and
archaeological layers into account, until the bedrock was reached. Archaeological
material from these levels was hand-sorted from 2/5 inch dry sieve residues, and
archaeological and geological information was documented on test-unit records.
In addition, at a few sites where units displayed unusual features or densities of
faunal material, samples for archaeozoological analysis were wet-sieved over 1
mm and 2 mm screens.
An auger testing campaign of Petite Terre's beaches had been planned for the
2000 field season, in order to test whether archaeological material would be present
underneath the beach sand and the dunes. Unfortunately, this could not be authorised
by the Office National des Forets within the available time frame as a result of
fieldwork limitations that apply to natural reserves.

Fieldwork Constraints

The fact that the islands of Petite Terre are uninhabited and strictly protected as
natural reserves greatly limits the possibilities for archaeological investigations. As
mentioned above, auger testing was not allowed on the Petite Terre beaches. In
addition, it is strictly forbidden to cut through the vegetation and to disturb the
populations of iguanas and birds that live on the islands. On Terre de Bas this severely
limited the possibilities for systematic and intensive fieldwalking. The fieldwork is
limited to short stays, as there is no drinking water on the islands, and without special
authorisations by the ONF and the mayor of La D6sirade, no overnight stays are
Topographic measuring of the test unit locations using GPS turned out to be
problematic as well, as a result of the relative scarcity of national reference points

and the relatively large distance between Petite Terre and La D6sirade. Two
geodetic engineers, who measured base lines toward three points on Petite Terre
from La D6sirade, solved this problem. Starting from those three points at Petite
Terre, short local base lines were measured from which the coordinates of the test
units could be positioned (Visschers and Lesparre, 2000).3

Fieldwork Results

The surveys on Petite Terre resulted in a site inventory that contains seven
sites (Figure 4). Bodu (1985) had previously reported five of these sites. Although
the 1998-2000 field surveys resulted in a very small number of new sites, they did
provide a much better understanding of the sites and their surroundings because
the sites and their environments could be mapped in detail and off-site material
could be studied as well.

Figure 4. Pre-Columbian sites at Petite Terre: 1. Baleine du Sud. 2. Site du Phare. 3. Est
de Mouton de Bas. 4. Mouton de Bas. 5. Trou Canard. 6. Pointe Sable. 7. Est de Trou

redicgad H-,


Habiton eftl ___ -m

The Petite Terre sites are single-component. Making rough chronological
assignments on the basis of stylistic and technological aspects of the ceramics
collected from the sites, the sites were dated to the Late Ceramic A period (AD

The functional assignment of some of the sites turned out to be a
complicated process. At the start of the surveys, it had been thought that
information on site types could be drawn from other, more extensive,
archaeological studies in the Caribbean. This turned out to be more complicated

than initially envisioned, because the rationale for functional assignments of
archaeological sites in the Caribbean often remains obscure in publications. More
importantly, the systematic and intensive nature of the East-Guadeloupe surveys
resulted in the discovery of a 'marginal' and under-exposed type of site,
characterized by shallow and dispersed distributions of fragmented and weathered
ceramics at the surface, which were not reported in detail in earlier studies in
Caribbean archaeology (De Waal, 1999b, 2002). At some of the sites lithic
artefacts and shell and coral fragments were found as well. The limited quantities
and the heavy fragmentation of the ceramics did not allow estimates of numbers
and shapes of vessels represented by the samples collected from the sites. This is
unfortunate since such estimates would have helped improve the understanding of
the function of these sites. It is thought that the indistinct sites may have been used
as temporary habitation sites, campsites or special activity sites that might have
been related to habitation sites (De Waal, 2006).
On Petite Terre, two such indistinct sites were discovered: the sites of
Mouton de Bas (12,800 m2) and Est de Trou Canard (4900 m2). They are both
located on Terre de Bas. The sites are situated on flat areas and they are close to the
coast, reefs, and soils that are suitable for small-scale horticulture. The sites are
characterized by small amounts of fragmented and weathered ceramics,
representing simple, open vessel forms, and shell fragments on the surface. The
sites appear to be shallow, but no units have been excavated to support this idea.
Small amounts of ceramic off-site material were found as well. The sites revealed
small quantities of material that are indicative of the local preparation of food,
such as griddle fragments and shell food remains. It has therefore been suggested
that people stayed at these specific spots for a while, at least long enough to
consider it necessary to prepare food on the spot instead of bringing it in a prepared
form. The sites may have functioned as satellite sites of Amerindian villages on
Pointe des Chiteaux or La Desirade. People living on Pointe des Chateaux and La
D6sirade may have been attracted to the abundance of marine resources around
Petite Terre (De Waal, 2006). According to Breton ([1647] 1978, 32),
pre-Columbian inhabitants of Guadeloupe went to Petite Terre to catch sea
turtles.4 It is not clear whether use of the indistinct sites was contemporaneous with
occupation of the Petite Terre settlements, but if it was, people using these sites
must have been in close contact with the inhabitants of Petite Terre, who depended
on the same marine resources as well.
The inhabitants of Petite Terre were living in large villages. The remains of
five such settlements were found (De Waal, 2006). The dimensions of the sites are

strikingly large, with site areas ranging between 8200 m2 and 25,100 m2. The
archaeological deposits are 60-80 cm thick. Since the village sites could not be
intensively excavated, it remains unknown how many houses were inhabited here
on a permanent basis. It is unknown as well how long the villages were occupied.
Large-scale excavations of the villages may solve those unanswered questions.

Bodu (1985) interpreted the Petite Terre sites as temporary camps that were
used for short stops on the way from Marie-Galante to La D6sirade and vice-versa.
This idea is probably linked to the often-voiced assumption that Petite Terre is too
small to be settled permanently and too dry to support human occupation.
However, archaeological material and site dimensions suggest that these sites
could very well have functioned as permanent habitation sites. Thick
accumulations of material have been found over large areas, and faunal and
ceramic assemblages appear to suggest permanent use. The idea that Petite Terre is
too small to allow permanent habitation appears to be guided by recent perceptions
of preferred living conditions.
Living at Petite Terre
Contrary to the opinion of many archaeologists, the islands were very
attractive for Amerindian habitation: they are surrounded by easily accessible
shallow water stretches, there are many canoe landing-spots and horticulture is
possible in almost all areas. The islands have an extremely high potential for
exploitation of marine resources, providing ideal living conditions for turtles, reef
fish and molluscs. This situation can be compared to the shallow water Saba-bank,
which is located 4.3 km southwest of Saba (Hoogland, 1996) and the
Anguilla-bank, which is close to the islands of Anguilla, St. Maarten/St. Martin,
and St. Barthelemy (Crock, 2000). Both shallow water banks are reported to have
been attractive for pre-Columbian habitants of the nearby region. On Petite Terre,
turtles today prefer to lay their eggs on the northern coast of Terre de Haut, and the
westernmost tip and at the southern coast of Terre de Bas (DIREN Guadeloupe,
1994, map in appendix 24). The Est de Trou Canard site and the Baleine du Sud
site are situated within a few minutes walk from these areas. In addition, the
islands appear to have harboured an abundance of seals (Breton [1665] 1892,
113).5 The seals mentioned were probably Caribbean Monk seals that are now
It cannot be denied that the islands are dry, however, one source of fresh
water was found during the 1998-2000 surveys. This source is situated near the
lighthouse, which is in the immediate surroundings of Site du Phare. As mentioned
above, fresh water lenses are probably present between chalky layers in the

subsurface in which pits can be dug to obtain drinkable water. The Amerindian
settlers of the islands probably had no shortage of drinking water, although some
effort may have been involved in collecting it.

The settlement sites yielded typical habitation refuse, including pottery,
shell and vertebrate food remains and lithic, shell and coral artefacts. The
ceramics, to begin with, are characterized by a large variety in vessel shape, which
is typical for village sites, as in settlements where a wide variety of activities were
carried out, all requiring a specific type and shape of pottery. Simple open, often
red slipped vessels (Figure 5), which may have been used for storage or cooking,
predominate the samples, however. Only a few of the vessels had been decorated,
seemingly emphasising the practical household use of the ceramics. Almost all
pottery had been finished by applying a red slip to embellish the vessels and, more
importantly, to make them less permeable to water.

Figure 5: Open vessel shapes from Est de mouton de Bas (A: unit 1, level 6, scale 1:4:
B:unit 1, level 2, scale 1:2: C: unit 1, level 2, scalel:3: D-F: unit 2, level 6).

%9bO 't



_ _. .




Samples of pottery from some of the sites show a large variety in rim shape,
surface colour, firing technique and surface finishing. This may suggest that
several potters were responsible for the manufacture of the pottery; each
household may have had its own potter. Apart from vessels, other household items
made of baked clay were found. These include fragments of griddles, which
demonstrate the local preparation of food, pot-stands, used for placing cooking
pots above a fire, and spindle whorls, indicating the local processing of cotton.
The inhabitants exploited shellfish (mainly Cittarium pica, Strombus gigas,
Nerita sp. and Acanthopleura granulata) that are easily and abundantly available
near the Petite Terre shorelines. The collected samples of faunal material reflect a
mix of everyday activities; in other words, no evidence for special or seasonal
activities that would be expected in temporary campsites has been found (Nokkert,
2006). People living at Petite Terre caught an abundance of terrestrial animals:
they ate a black land crab, great land crab, rice rat, iguana and birds. They also
seem to have enjoyed eating reef carnivores (shark, grouper, squirrelfish, jack,
snapper, grunt, wrasse and Spanish hogfish), reef herbivores/omnivores
(parrotfish, triggerfish and surgeonfish), and inshore species (red rock urchin,
needlefish, common spider crab, sea turtle and porgy). Off-shore pelagic species
(tuna/mackerel and barracuda) were less regularly caught. In addition, they
probably ate the crops from their horticultural plots and they may have added wild
fruits, tubers, leaves and seeds to their diet.
The pre-Columbian inhabitants of Petite Terre used artefacts made from
rocks that were obtained from La Desirade. These include water-worn pebbles,
grinding stones, and core-artefacts. A hammer-stone fragment made from possible
Basse-Terre (Guadeloupe) hypabyssal rock was found as well. They also obtained
rocks from more distant sources, such as flint from Long Island and Antigua, and a
ground petalloid axe of possible St. Martin/St. Barths chert. Local stone materials
were used as well, for example, for the manufacture of a carved limestone zemi
(COLOUR PLATE 1, p. iv), non-modified flake artefacts and core artefacts.
Shell artefacts were only found in two of the village sites. These include a
few beads (Murex sp. and Oliva sp.) at Site du Phare and a polished axe fragment
and a polished and decorated spatula (both of Strombus gigas) and a carved,
polished and incised Cittarium pica fishhook at Est de Mout2on de Bas
(COLOUR PLATE 2, p.iv ).
Coral, abundantly present close to the Petite Terre shores, was intensively
used for rasps (Acropora cervicornis) and passive grinding tools (Acropora
palmata). Rasps and grinding tools may have been employed in the processing of

vegetable foods, pigments, as well as in the manufacture and finishing of shell
artefacts. Most of the coral artefacts had been heavily used and had, as a result,
severely flattened and ground edges.

Finds of a shell spatula (Est de Mouton de Bas; Plate 2, p.iv), a human face
bowl (Pointe Sabl6) and a bird adorno, a figurine leg fragment and a zemi from
Site du Phare (see Plate 1, p.iv) suggest that activities of a more spiritual nature
took place in the villages as well.
When trying to understand the ways in which pre-Columbian Amerindians
were living at Petite Terre, it is important to note that they may have considered the
islands to be far from isolated, as we tend to see them today. The sea surrounding
the islands probably functioned as a transportation route for people, utilitarian
goods and raw materials, foodstuffs, technological innovations and ideas and
beliefs. Regular contacts with inhabitants of La D&sirade, Marie-Galante and
Grande-Terre (Guadeloupe) or even with people living on more remote islands
would have greatly enhanced the pre-Columbian living environment on Petite

The settlement history of Petite Terre closely follows that of other areas
investigated during the East-Guadeloupe surveys (De Waal, 2006). No evidence
has been found for habitation of Petite Terre, Pointe des Chateaux and La Desirade
during the pre-ceramic and the Early Ceramic A periods (2000 BC-AD 400).
There are no indications that people lived on Petite Terre during the Early Ceramic
B (AD 400-600/850) either. But during this period, pre-Columbian Amerindians
did live in six large permanent coastal settlements at Pointe des Chateaux and on
La Desirade.

People started inhabiting and exploiting Petite Terre during the Late
Ceramic A period (AD 600/850-1200/1300) for the first time. This coincides with
an increase in the number of villages in Pointe des Chateaux and La D&sirade. For
this period we can thus try to reconstruct the micro-regional settlement pattern and
by doing that, try to understand the way the Petite Terre settlements may have
functioned in a micro-regional or regional perspective.
For the Late Ceramic A period, a total of 22 settlements, two ceremonial
locations, one strategic outpost and 34 indistinct sites were found in the
investigated micro-region consisting of Pointe des Chateaux, La D6sirade and
Petite Terre (Figure 6). Five of these settlements and two indistinct sites are
located on Petite Terre.

Occupation of the micro-region was concentrated in large permanent
villages. Starting during the Early Ceramic B period, occupation of the
micro-region obviously consolidated and intensified during the Late Ceramic A
period. It is not yet clear how this increase in numbers of settlements for the Late
Ceramic A period can be explained. It is possible that the population grew and thus
more settlements were needed. It is also possible that settlements were inhabited
for shorter periods and moved to previously uninhabited locations from time to
time. A third possibility is that inhabitants of increasingly densely occupied
villages decided to split up their growing community into smaller social groups
and to found new, smaller, economically independent settlements. By doing this,
they might also have effectively filled up the open spaces within the landscape. As
a result, the distances between the settlement and the settlement territories were
smaller compared to the preceding Early Ceramic B period. Most settlements
probably had territories of roughly similar size.
During the Late Ceramic A period, as in the Early Ceramic B, most
settlements were still situated along the coast and the inhabitants exploited zones
near the settlements. People practised a mixed economy, focusing on root crop
horticulture, hunting of land animals, fishing, catching birds and collecting
molluscs, fruits, wild tubers and seeds. The diversity in site types increased and a
greater proportion of the landscape was used for ceremonial, socio-political and
economic activities at La D6sirade in particular. It is obvious that during the Late
Ceramic A period, the landscape was used intensively. On La D&sirade, special
ceremonial sites were established, as well as temporary camps and sites related to
gardening activities and to the exploitation of natural resources outside the
villages, and caves were used for shelter. On Petite Terre, two indistinct sites were
interpreted as satellite sites of Pointe des Chateaux or La Desirade villages, aimed
at the exploitation of marine resources that abound around Terre de Bas and Terre
de Haut.
Inhabitants of some of the Pointe des Chateaux and La D6sirade villages
participated in long distance contacts, stretching as far as the South American
mainland and the Greater Antilles. These contacts were less frequent than during
the Early Ceramic B, but they continued to exist. The inhabitants of Site Du Phare
on Petite Terre also participated in regional contact networks in which finished
green chert artefacts circulated. These objects were exchanged as finished items
after they had been manufactured in villages in St. Martin or St. Barths
(Knippenberg, 2006). Inhabitants of the Site du Phare, Est de Mouton de Bas and
Pointe Sable villages also obtained flint from Long Island and Antigua and they

undertook trips to La D6sirade and Basse-Terre (Guadeloupe) to collect volcanic
rocks for the manufacture of tools as well (De Waal, 2006).

During the Late Ceramic B period (AD 1200/1300-1493) the islands of
Petite Terre become deserted again. This coincides with a dramatic decrease in site
numbers in the micro-region: only two small villages on La D6sirade and one
small village on Pointe des Chateaux were inhabited during the latest part of the
pre-Columbian period. This depopulation may have been an effect of the influence
of developing cacicazgos on the Greater Antilles, attracting people from the
Lesser Antilles because of the greater social opportunities. The Late Ceramic A

Figure 6 : Late ceramic A sites in the Eastern Guadaloupe micro region

II -,


S 12*.'13 17

A Ceremonial site
* Indistinct te
* Settlement
4 Strategic outpost

0 ---
00 1
1 k

population increase-and possibly a related decline in the availability of food
resources-may have provoked socio-political dissatisfaction amongst people
living in the Lesser Antilles and made them susceptible to the attractions of
complex social societies on the Greater Antilles. The islands of Petite Terre would
not be inhabited again until the eighteenth century, when European settlers started
to use the islands for the production of cotton.


Reviewing the data presented, it is striking that archaeological evidence for
pre-Columbian habitation and exploitation of the islands of Petite Terre dates only
from the Late Ceramic A period (AD 600/850-1200/1300). During the Early
Ceramic B and Late Ceramic B periods, in which surrounding islands were
inhabited, these small islands were desolate. This abandonment is especially
remarkable in the light of the considerable possibilities for settlement and resource
exploitation offered by the marine environs of the islands.

A second result of this fieldwork is that the Petite Terre sites have long been
interpreted as temporary camps, used for short stops on the way from
Marie-Galante to La D&sirade and vice-versa. However, archaeological
investigation of the sites and study of the materials collected suggest that the
islands were inhabited permanently during the Late Ceramic A period, as they
were during colonial and more recent times as well. The Petite Terre villages are
surprisingly large. The islands provided sufficient natural resources to allow
permanent habitation and, as archaeological investigations have demonstrated,
people inhabiting the islands were involved in contacts with people on other
islands thus greatly enlarging their social, political and economic environment.
The settlement of Petite Terre seems to mirror the developments in settlement
pattern, social organisation and interaction that were also witnessed at Pointe des
Chateaux and La D6sirade (De Waal, 2006).

Our present-day perceptions of what are preferable living conditions should
not guide us in determining the functions of pre-Columbian sites. Small and dry
islands, such as the islands of Petite Terre, can definitely harbour large permanent

1. The archaeological research on which this article is based was financed by Leiden
University, the Netherlands, and the French Ministry of Culture. The fieldwork was
conducted by F. Asselbergs, M. Dorst, M. Kappers, K. Leijnse, J. Lesparre, E. Mietes, A.
Miller, M. Nokkert, F. Stevens, M. Van de Heiden, W. Visschers and the author and a
lot of help was offered by members of the Anse I la Gourde (Guadeloupe) excavation

teams. Mr. Robin, former mayor of La D6sirade, and Mr. Maston, director of the land
registry office, provided the special authorisations required. S. Knippenberg analysed the
lithics, M. Nokkert studied the vertebrate faunal remains and J. Lesparre and W.
Visschers provided co-ordinates of the test unit locations. Figures were made by E. Van
Driel artefactt drawings), M. Oberendorff (maps) and J. Pauptit (photographs).
2. The 2000 fieldwork consisted of topographic measuring of the test unit locations
3. The location of each test unit was measured using two Leica SR261 GPS-receivers
and two Leica CR344 field manuals and a Sokkia SET 4B infrared theodolite.
4. "l y a autour de la Guadeloupe nombre de petites fles fort agreables. [...]
Marie-Galande et La Desirade n 'en sont pas loin ny la Petite Terre esquelles nous allons
aussy dans la saison pour turner les tortues".
5. "Cay6oli: Islet entire la desirade & la pointe de la grande terre appellee
premierement la petite terre du depuis I 'islet aux ours marines, enfin I 'islet d 'hoiiel".

Antczak, A. (1998). Late prehistoric economy and society of the islands off the coast of
Venezuela: A contextual interpretation of the non-ceramic evidence. Unpublished PhD
dissertation, University College of London, London.
Barr6, N., Lorvelec, O. & Breuil, M. (1997). Les oiseaux et les reptiles des miles de la
Petite Terre (Guadeloupe): Bilan d'un suivi 6cologique d'une ann6e (mars 1995 a mars
1996). Rapport de I'Association pour 'Etude et la protection des Vertebres des petites
Antilles (AEVA), no. 16. Unpublished typescript, AEVA, Petit-Bourg.
Bodu, P. (1985). Rapport des prospections en surface. Desirade. Unpublished typescript,
Musee Edgar Clerc, Le Moule.
Breton, R. ([1647] 1978). Relations de l'ile de la Guadeloupe, Tome 1. Basse-Terre,
Guadeloupe: Soci6te d'Histoire de la Guadeloupe.
Breton, R. ([1665] 1892). Dictionnaire Caraibe-Francois. Facsimile edition, Leipzig:

Conservatoire du Littoral (1997). Plan de gestion des ilets de Petite Terre: 1997-2001.
Unpublished typescript, Conservatoire du Littoral, Pointe-a-Pitre.
Crock, J. G. (2000). Inter-island interaction and the development of chiefdoms in the
Eastern Caribbean. Unpublished PhD dissertation, University of Pittsburgh, Pittsburgh.
De Waal, M. S. (1998). Field manual. The Pointe des Chdteaux and Petite Terre survey
program (1998). Unpublished typescript, Leiden University, Leiden.
De Waal, M. S. (1999a). Field manual. The La Desirade and Petite Terre survey (1999).
Unpublished typescript, Leiden University, Leiden
De Waal, M. S. (1999b). The Pointe des Chateaux survey (1998). A preliminary report.
In G. Richard (Ed.), Proceedings of the XVIIth International Congress for Caribbean

Archaeology, Part 2 (pp. 268-276). Guadeloupe: International Association for Caribbean
De Waal, M. S. (2002). Occupations Amerindiennes dans I'Est de I'Archipel
Guadeloupeen. Rapport de Synthese. Tome I: Recherches de Terrain 1998-2000; Tome
II: Inventaire du Mobilier Decouvert; Tome III: Annexes. Leiden: De Waal Publishers.
De Waal, M. S. (2006). Pre-Columbian social organisation and interaction interpreted
through the study of settlement patterns: An archaeological case-study of the Pointe des
Chdteaux, La Desirade and Les Iles de la Petite Terre micro-region, Guadeloupe, F. W.I.
Leiden: De Waal Publishers.
DIREN Guadeloupe (1994). Dossier de creation de la Reserve Naturelle terrestre et
marine des Ilets de Petite Terre. Unpublished typescript, Office de Genie Ecologique,
Hoogland, M. L. P. (1996). In search of the native population ofpre-Columbian Saba
(400-1450 A.D.). Part two: Settlements in their natural and social environment.
Unpublished PhD dissertation, Leiden University, Leiden.
Knippenberg, S. (2006). Stone artefact production and exchange among the northern
Lesser Antilles. PhD dissertation, Leiden University, Leiden.
Nicholson, D. V. (1975). A possible historic Indian site on Petite Terre, Guadeloupe.
Unpublished typescript, Antigua Archaeological Society, Antigua.
Nokkert, M. (2006). Appendix 5: Faunal remains from the pre-Columbian sites of Pointe
des Chateaux, La D6sirade and Petite Terre. In: M. S. De Waal, Pre-Columbian social
organisation and interaction interpreted through the study of settlement patterns: An
archaeological case-study of the Pointe des Chdteaux, La Desirade and Les lies de la
Petite Terre micro-region, Guadeloupe, F.W.I., 363-410. Leiden: De Waal Publishers.
Rousteau. A. (1995). Etude phytoecologique des ilets de Petite-Terre. Basse-Terre,
Guadeloupe: Office National des Forets.
Visschers, W. & Lesparre, J. (2000). Archeologisch onderzoek Guadeloupe: Inmeten
testputten met GPS en total station. Unpublished typescript, Afdeling Geodesie,
Technische Universiteit Delft, Delft.
Watters, D. R. (1980). Transect surveying and prehistoric site locations on Barbuda and
Montserrat, Leeward Islands, West Indies. Unpublished PhD dissertation, University of
Pittsburgh. Ann Arbor: University Microfilms.

Targeting the Jamaican Ostionoid: The Blue
Marlin Archaeological Project'


Status of Jamaican Prehistory and Redware Studies
The study of Jamaica's earliest prehistory has yet to receive concentrated
and widespread attention from archaeologists. One reason for this is that, until
recently, too few professional archaeologists have existed on the island, and fewer
still have been tempted by, or equipped for, the study of the island's prehistory.
This situation is changing gradually as university-trained archaeologists continue
to emerge in Jamaica, nonetheless, such a positive development is counteracted by
a persistent negative: professional archaeologists outside Jamaica have generally
not addressed the island's rich prehistoric resource. Thus, within the entire
Caribbean, Jamaica remains one of the least known islands prehistorically. Even
within the Greater Antilles, knowledge of Jamaica's pre-Columbian peoples lags
behind that of Cuba, Haiti, the Dominican Republic, and Puerto Rico.
The work of the Blue Marlin Archaeological Project on Jamaica's south
coast (Figure 1)2 is seeking to address the paucity of knowledge concerning the
earliest known occupation of Jamaica, the Ostionan Ostionoid culture,3 informally
referred to as 'Redware.' This culture predated the arrival (on Jamaica) of the
better-known Taino (Meillacan) prehistoric populations by about two hundred
years. The present archaeological remains of both cultures show that Jamaican
Redware and Taino peoples were two distinct cultural entities, at least in their
early stages, although it is suspected-but not proven-that the Redware culture
became assimilated into that of the Taino through a process of social and cultural
interaction over time. Thus, the Redware culture could well have evolved into
what we now recognize as Taino, without the intermediate stages of such
development yet apparent in the thin archaeological record. At its outset, the
Redware culture is distinct from the Taino culture by the characteristics of its
ceramics, its habitation preferences, and by its dietary habits. For example, the
presence of red slipped or red painted ceramics in Redware assemblages (hence
the term Redware) is not known in Taino assemblages. These ceramics bear a fine
sand temper in contrast to unslipped and unpainted Taino wares with coarser
tempers. The settlement pattern of Redware peoples at present reflects a coastal

orientation where people lived within 100 metres of the shoreline, while the later

Taino populations generally preferred inland locations on high ground

overlooking the sea, although some coastal Taino sites are known on Jamaica.

Figure la and Ib: Location of Blue Marlin site, 60 m from the sea

Fig la


%MANCHESTER", ANDREW ..............


Fig lb

Only two radiocarbon dates (Vanderwal, 1968, p. 130; Keegan, 2000a, para. 2)4
have been obtained for the Jamaican Redware, which presently place this cultural
sequence between c. A.D. 650 and c. A.D. 850.5 The Taino culture is known by a
more sizeable collection of radiocarbon dates from a number of sites, which place
it between A.D. 800 to A.D. 1494. Thus, it was the later prehistoric peoples, the
Tainos, whom Columbus met when he arrived on Jamaica, with all visible traces of
the Redware peoples having disappeared by that time. The sparse archaeological
record so far supports the disappearance of the Jamaican Redware culture within
prehistoric time, and thus, Redware peoples were unseen, unaffected, and
unrecorded by historic populations.
At the time of writing, no earlier culture, either ceramic or ceramic,
predating the Redware sequence has been reported on Jamaica. It is likely that this
circumstance is not an accurate reflection of prehistoric reality, since both ceramic
and ceramic complexes have been found in pre-Ostionoid sequences on
neighboring islands of the Greater Antilles (Cuba, Hispaniola, Puerto Rico).
Concerning the earliest occupation of Jamaica, one writer notes:

It is quite remarkable that such a large land mass in the Americas
went uncolonized for so long. It may well be that evidence of
earlier occupations or at least visits will eventually come to light,
for Jamaica is just beginning to receive the intensity of
archaeological research that the other Greater Antillean islands
have experienced (Wilson, 2007, p. 102).

What is more remarkable is that the date of the earliest migration to the Greater
Antilles (from Central America), which populated this chain of islands, took place
as early as c. 4000 B.C., leaving Jamaica apparently untouched.6 Scholars
working in Caribbean prehistory have advocated the possibility of pre-Ostionoid
sites on Jamaica so frequently (e.g. Reid, 2009, p. 76; Wilson, 2007, p. 102) that its
likelihood must now become a working assumption if the knowledge of Jamaican
prehistory is to advance. The writer proposes, therefore, that pre-Ostionoid sites
on Jamaica must be properly targeted through a program of systematic and
intensive survey if they are to be located. No comprehensive archaeological
survey of the Jamaican coastline has yet been undertaken, and this gap needs to be
addressed within the context of Ostionoid archaeology. A survey of this type must
be both archaeological and geological, taking into account the probable loss of
coastal sites through land subsidence under the sea. Redware sites along the north

coast, known about forty years ago have already been lost in this manner (see table
Distribution of Jamaican Redware Sites
It is difficult to arrive at numbers of distinct Ostionoid sites in Jamaica,
because of the tendency of scholars to count numbers of sites (in publication),
without listing sites specifically by numeric and/or named designations. Including
the new Blue Marlin site, nineteen Redware sites are known for certain, and these
are listed in table 1. Most are habitation sites, but a few are caves used in
prehistoric time for burial. An additional eight sites are mentioned by Lesley-Gail
Atkinson (2006, p. 130), but their location and their nature can not be determined
from the existing published literature. In table 1 the 'Lee designation' refers to the
island-wide coding of sites by parish by James Lee between 1960 and 1980, and is
included here for the benefit of those familiar with this site reference system. For
others, it is better to go by the name of the site, although some names such as
'Alligator Pond' have been used more than once to refer to different sites in the
same vicinity. Because of this, Lee's alphanumeric system is still useful for
distinguishing those sites with duplicated place names discovered before 1980.
If the corpus of sites is plotted on a map, the result is that three distinct
spatial clusters of Redware occupation emerge along the Jamaican coastline. One
cluster occurs on the northern coast of St. Ann parish, a second in the southern
adjacent parishes of Manchester and St. Elizabeth, and a third along the
southwestern coastlines of Westmoreland and St. Elizabeth (Figure 2). I suggest
that such a widely spaced distribution is 'suspicious', and is a good indication that
Redware sites might well be found all around the island's coastline if an intensive
systematic survey of the type proposed above were undertaken.
Other than the Blue Marlin site, only Paradise Park in Westmoreland has
undergone systematic excavation in one season, with an additional season of test
survey having been undertaken (see Keegan, 1998 and 2000a for reports). At
Paradise Park, indications are that both Redware and Taino peoples inhabited this
locality, albeit in spatially distinct contexts. Recent analysis of shell remains from
Paradise Park (Keegan et al. 2003) has begun to yield comparative information
about how Jamaica's two prehistoric populations might have adapted differently
within closely similar environments.

Table 1: Jamaican Redware habitation sites and burial caves

Site Name e Location Status/notes Publication
St. Ann parish, 1963, by Vanderwal,
1. Alloa A-4
north coast highway 1968: 96

Type site for
de Wolff,
St. Ann parish, culture.
2. Little River A-15n lture Vanderwal,
north coast Destroyed;
1968, 94, 96;
washed into
Lee 1980, 597
St. Ann parish, Destroyed by
3. Runaway Bay A-30 Lee, 1976, 2
north coast construction

One of 12 sites
None; not a St. Ann parish, Atkinson,
4. Mammee Bay since 1980,
Lee site north coast 2006, 130
after Lee's
Storm damage
St. Elizabeth has resulted in
5. Alligator Pond E-1 parish, south disturbed and '
1968, 96, 99
coast unstratified
St. Elizabeth Destroyed by
6. Black River E-3 parish, south building
1968, 100
coast activity
St. Elizabeth Completely 1968, 99; Lee,
7. Great Pedro Bay E-4 parish, south covered by a 1976, 3; Lee,
coast sand dune 1979, 4; Lee
1980, 601
Storm damage
St. El h has resulted in
St. Elizabeth
disturbed and Vanderwal,
8. Alligator Pond E-5 parish, south
unstratified 1968, 96
deposits as for

A Redware
St. Elizabeth A Redware
9. Parchment Cave EC-5 parish, south cavepotter Lee, 1979, 1,4
& skeletal
coastmaterial found

St. Elizabeth Status
parish, SW unknown.
coast, between Excavated by
10. Longacre Point E-6 ot eten te Lee, 1976, 3
Luana Point Brother
and Black Michael in
River 1966
St. Elizabeth
11........ E-6 parish, south unown Lee, 1976, 3

Yielded turtle
St. Elizabeth
EC-10 (burial fa bowl and Lee, 1979, 1,
12. Baalbec parish, far
cave) i d skeletal 4-5
St. Elizabeth Lee, 1976, 3;
13. Calabash Bay E-ll parish, south unknown Lee, 1979,3;
coast Lee,1980 601
1 km from
St. Elizabeth
shore on solid
14. Sandy Bank E-13 parish, south sh ons Lee, 1979, 1
sandstone, not
St. Elizabeth
SNone: not a parish, near Good Rampersad,
15. Blue Marlin
Lee site Great Bay, condition this article
south coast

land parish, Ebanks, n.d.;
Designated W SW coast. One
-11 but not by of 12 new sites Good eegan,1998
16. Paradise Park & 2000a
iLee (see discovered condition
Ebanks 1992) since 1980 Atkinson,
after Lee's 2006, 130

One of 12 new Status
sites unknown.
None: not unknown. Atkinson,
17. Anderson e t discovered (Near
Lee site 2006, 130
since 1980 Alligator
after Lee's Pond, M-4)

Partially or
Manchester wholly
18. Bottom Bay, arh, soth o y 1968, 96, 99;
parish, south covered by
a.k.a. Alligator a.k.a. M-4d b Lee, 1976,3;
Pond coast; 180m sand dunes. Lee,1979,2;
Pond Lee, 1979, 2;
from sea Excavated by
Lee, 1980, 597
One of 12 new
19. Porus None: not a i Status Atkinson,
Manchester Lee site 18 unknown 2006, 130
since 1980
after Lee's
Eh undetermin-
additional sites tAtkinson,
20-27 able from Uunknown 200
found after 2006, 130
1980. published
1 literature

The Blue Marlin Site

The Blue Marlin site is located at latitude 170 52' 1.81" N and longitude 770
44' 25.77" S. The area designated an archaeological site (see Figure Ib, white
outline) is an undeveloped portion of a two-acre beach resort property on the south
coast of Jamaica in the Great Bay/Treasure Beach region. As a prime resort area,
the property surrounding the site is occupied by holiday cottages and other guest
amenities, and in addition, plans are underway for the future development of the
area now designated an archaeological site. This area (Figure Ib, white outline), is
presently fenced off from the remainder of the property, and it is this factor that has
determined the site's archaeological boundaries, since the rest of the property is
off limits for field investigation. With development plans underway, time is of the
essence in the investigation of this site, but it should be noted that the landowners
are keenly interested in developing the archaeological resource on their property;
they have in fact granted permission for its investigation over a three-year period.

Fig 2: Distribution of Redware sites in Jamaica


.. . ST AST < A RY
WESTMOR ,A--.~t -


Blue Marlin

I 100km

Until 2007 no formal archaeological investigation had yet been undertaken on any
part of the property, although it has long been known from the avocational
interests of the former landowner, who undertook limited, informal excavations,
that the two acres of land is rich in Redware cultural remains. In addition, years of
surface collecting by the landowners, past and present, have yielded an impressive
private collection of prehistoric artifacts (ceramics and stone tools), although,
unfortunately, none of this material has yet been published.

In January 2007, the writer first visited and confirmed that the Blue Marlin
property did belong to the small corpus of sites in Jamaica that contains Redware
material. This assessment was informed by a fieldwalking survey of the fenced
perimeter or site, which yielded surface finds of Redware sherds as well as small
amounts of prehistoric lithic material. Most importantly, I have determined by
now that the Blue Marlin property was not listed in any past archaeological
surveys, particularly those of Dr. James Lee. It is not to be confused with the E-4
site at Great Bay (see table 1), near the Blue Marlin property and already noted by
Lee. The Blue Marlin site is therefore a new addition to the known corpus of
Jamaican Redware sites (table 1). The first formal archaeological excavation
began there in August 2007.

The site's location within 60 m of the coastline fits the general profile given
by Lee for Redware sites in Jamaica (Lee, 1979, p. 2). Its proximity to the beach, as
the faunal material will show (below), no doubt facilitated the exploitation of
coastal food resources. Also consistent with Lee's profiling is a shallow cultural
deposit of only 30-45 cm deep, making it typical amongst the known Redware
deposits on the island. The site is both shallow and stratigraphically

uncomplicated, with two brief layers of cultural deposition immediately overlying
the virgin subsoil. Minor amounts of historic material (British colonial) were
recovered within the top few centimetres only, but for the most part the site is
'pure' Redware with no evidence of a later Taino occupation within or around the
excavated area.

The 2007 field season has established that the Blue Marlin site was certainly
a place of habitation for at least a small population of Redware people. In the single
(2 m x 2 m) trench opened in 2007 a compacted floor surface of a prehistoric
structure was uncovered along the eastern borders of the unit, indicating the
presence of an ancient settlement in the northern portion of the site. Three small
postholes (8-10 cm in diameter) were also found in the basal levels of the trench,
forming a partial circular patterning underneath the floor and thus indicating the
presence of a rounded structure. Sectioning of these three features unfortunately
yielded little material evidence, however, by their small and shallow dimensions, it
is assumed that a small or light-framed structure is represented. Further efforts to
uncover the entire floor surface and its associated postholes were left unfinished
until the 2008 season, when it is hoped more of the feature can be found,
excavated, and recorded.

The Blue Marlin Ceramics

Within the short two-week excavation season (cut short by hurricane Dean),
2,207 ceramic sherds were recovered and 220 or ten per cent found to be
diagnostics (rims, bases and handles). Such a small number of diagnostic sherds is
not considered representative of the entire pottery assemblage of the population
that once occupied this site, and it will take at least another season of data
collection to acquire a more representative sample. Nonetheless, the diagnostics
recovered so far are being formally drawn and used to begin a typological
sequence of forms. In the absence of formal typological studies on Jamaican
redware, the justification for using this type of classificatory system is to arrive at a
definition of Jamaican redware by documenting its forms and eventually its fabric
types. It is hoped that in time absolute dates can be assigned to specific forms and
fabrics and that Jamaican redware will be able to stand, first, on its own within its
internally defined chronology, and secondly within the wider comparative context
of Ostionoid sequences from other islands of the Greater Antilles.

The pottery is hand-made (not wheel-made), thin, hard, smooth on both
surfaces, very well fired, and thus typical of most other Ostionoid assemblages
throughout the Greater Antilles. The fine texture sets it apart from the much

coarser Taino wares, supporting the consistent observation of a more superior
ceramic tradition amongst the earlier (pre-Taino) ceramic-using peoples of the
Greater Antilles.
The Blue Marlin forms thus far show a preference for inwardly curving
bowls and pan-like vessels of a variety of sizes (Figure 3), although straight-sided
open forms are present among some smaller vessels (Figure 3, lower).
All are likely domestic forms intended for food preparation, storage, or
eating purposes. The most striking vessels are undoubtedly the handled wares of
either round or elliptical shape (Figure 4). From studies of Puerto Rican Redware
ceramics, loop-shaped handles as seen here are a feature of early Ostionoid
sequences (Espenshade, 2000, p. 4), and are a distinguishing feature of the early
Ostionoid expansion within the Greater Antilles. Other features of the Blue Marlin
Figure 3: Domestic wares: Large pans and bowls

/ Cm
/ I
( I m 5cm

( \ -----
r~- r----

ceramics that suggest an early Ostionoid sequence are the relative rarity of
decoration on vessels, and the low occurrence of burnished or semi-burnished

Figure 4: Handled cooking wares and decorated sherd

m 5 cm

S-..... r .. -


Blue Marlin Shell Industries
It is generally well understood that prehistoric peoples in the Caribbean
exploited shell resources as a food source, but more significant for the Blue Marlin
site is the use of shell as tools. There is evidence that a number of different tool
types were made on a select variety of shells, notably Strombus gigas (queen
conch), Codaika orbicularis (tiger lucine), and Cittarium pica (West Indian top
shell). Tool types identified from the 2007 shell assemblage include

(1) gouges (Figure 5), including one attractive incised example (Figure 6)
(2) spoons or scoops made from the lip of a conch (figure 7),

(3) possible hammers made from entire juvenile conch, showing beveled tips
from heavy use (Figure 8),

Scrapers made on tiger lucine and West Indian top shell (Figures 9 and 10

Planing tool made from a half-sectioned body of conch shell (Figure

Possible drills made from the internal columella of the queen conch
(Figure 12).

Figure 5: Gouges made from conch

Figure 6: Decorated (incised ) tool from conch

s s s a cm

Figure 7: Strombus gigas scoops and spoons

Figure 8: Strombus gigas hammers

Figure 9: Codakia orbicularis scraper

Figure 10: Cittarium pica scraper

Figure 11: Strombus gigas plane

Figure 12: Strombus gigas, possible drills

From analyses of contemporary shell assemblages on other islands of the Greater
Antilles (O'Day & Keegan, 2001) it is possible to conclude that the shell
implements found at the Blue Marlin site are 'expedient' (O'Day & Keegan, 2001)
rather than formal tools; that is, the shell was merely broken or cut into the desired
shape and without further reduction or other modification was used for the task(s)
required. This impromptu use of shell tools has important implications for the
economic interpretation of this culture. Clearly a minimum amount of energy was
expended in the production of shell tools, which might be a function of the high
availability of this raw material within the environment. There would have been
little need to curate (keep, maintain, and reuse) tool forms for long periods of time,
and when forms became worn or broken, new forms could be produced as
replacements readily and quickly. Much more needs to be said about this important
class of evidence. Determining tool use from any material type is a more involved
process than simple observation with the naked eye. Rather, one requires
microscopic use-wear analysis of the working edges in order to define patterns of
wear that might indicate use on specific materials, such as wood, stone, etc.
Use-wear analysis, therefore, has the potential of determining likely tasks
employed by the shell implements recovered from the site, and this is the direction
in which further analysis of this tool kit is likely to proceed.
Significantly, some shell types seem never to have been used as tools, and
their appearance on the site in large numbers attest to their exploitation as a food
source only. These species include Area zebra (turkey wing), Donax denticulatus
(common Caribbean donax), Fissurella nodosa (knobby keyhole limpet), and
various varieties of Pleurodonte or land snail.

Faunal Remains
A small faunal assemblage of about 350 bone fragments has been recovered
thus far from the site. This material has undergone preliminary analysis, with the
following broad groups emerging:7

Pisces 7%
Cheloniidae/Trachemys terrapen 77%
Aves 0.9%
Geocapromys brownii 4%
Unidentified 10.4%

Clearly the overwhelming majority of the bone material is that of turtle (with
exact species yet to be confirmed), indicating a heavy exploitation of this class of
fauna. Cheloniidae represents the family of sea turtles, while Trachemys terrapen
is the Jamaican Slider Turtle, a freshwater species endemic to Jamaica and
presently endangered. It still inhabits the wetter regions of the southern coast of
Jamaica as in prehistory, in environments ranging from permanent and seasonal
ponds, to caves that hold water during the dry season, as well as streams. It is not
known to inhabit the mangrove regions of the island. It is most interesting that at
Paradise Park, turtle remains also dominate the faunal assemblage, although the
species and their exact proportions have yet to be reported.
The remains of bird (Aves) are thought to be a natural occurrence in the site
and not necessarily a result of cultural activity, since a few bones only were
uncovered at the top levels of the subsoil. Given the coastal orientation of the site,
it is surprising that so little fish remains have been recovered, despite careful dry
sieving (1/8-inch mesh size) to collect such remains in the field. Similarly, the low
recovery of Geocapromys brownii or the Jamaican hutia (coney) is equally
surprising, as this is a species known to have been more heavily exploited by the
Taino prehistoric populations on the island.
Assessment of the Blue Marlin Site

The shallowness of the habitation remains at the Blue Marlin site indicates
one or more factors: a small population living on the site, a short period of
occupation, and/or a seasonal occupation based on the availability of desired
marine resources. It is yet to be determined whether the site might have been a
seasonal rather than a permanent settlement, however, the recovery of more faunal
and shell species will be instrumental in determining this, as will a better definition
of the structure type and construction. Indications are, from other Redware

complexes in the Greater Antilles, that these people were not nomadic or even
semi-sedentary, and that they did settle in permanent communities. The
shallowness of their habitation sites, also observed outside of Jamaica, has been
explained by the rapid expansion of Ostionan groups out of Puerto Rico and across
the islands of the Greater Antilles, including Jamaica (Keegan, 2000b, p. 150).
Thus the speed at which the Ostionan expansion took place could well have
resulted in "settlements that were small and widely scattered" (Keegan, 2000b, p.
150) throughout the northern Caribbean.
Economically there is little evidence for hunting as a means of subsistence at
the Blue Marlin site, as attested by the virtual lack of land mammals (Jamaican
coney only) and their small percentage. Small flakes of chert, which are likely
indications of hunting activity, have been found in this site, but their numbers are
minimal at best. This material is undergoing analysis, but its scarcity does not
support extensive hunting pursuits. As indicated, this contrasts sharply with the
Taino inland populations who exploited a larger proportion of the Jamaican coney,
and whose sites exhibit a more abundant lithic repertoire than at the Blue Marlin
site. Furthermore, there is a de-emphasis of the fishing economy at Blue Marlin, in
contrast to the well-represented fishing complexes in the later Taino sequences on
the island. Instead, we see an overrepresentation of the sea turtle, suggesting that,
in addition to shell foods (and perhaps in preference to shell foods), turtle was a
main source of protein in this population's diet. Although little can be said yet
about the nature of procurement of food at this site, one need not assume a
knowledge of deep sea fishing for the exploitation of sea turtle; females of most
species would have been easy prey when they ventured onto shore to lay their
eggs, particularly between May and October as they migrated to their annual
nesting beaches.8 At this time they could be caught easily with the bare hands by
overturning them onto their backs.9 In addition, we can assume that eggs, when
deposited by turtles onto dry land, were harvested for food by Redware peoples. 0
although this aspect of subsistence would leave little trace in the archaeological
record. A seashore type of exploitation such as this, tied into the reproductive
cycle of a species, would tend to support a seasonal occupation for the Blue Marlin
site. We should therefore entertain the theory that the shallow nature of Redware
sites in general and of the Blue Marlin site in particular might be tied into a regular
seasonal exploitation of the sea turtle as a main food supply for these early
prehistoric peoples. More remains to be said about Jamaican Redware turtle
exploitation once exact species of these reptiles are identified through formal
faunal analysis.

A shoreline subsistence pattern for the Blue Marlin culture is further
supported by all of the shell species encountered thus far, which are inter-tidal,
shore oriented, and/or seagrass varieties. Mangrove exploitation is not represented
here, which again contrasts sharply with known Taino patterns of subsistence. The
Redware preference for seagrass exploitation and the Taino preference for
mangrove swamps have been suggested through the shell assemblages at Paradise
Park (Keegan, Portell, & Slapcinsky, 2003). A comparison of mollusc species
from both settlement sequences at Paradise Park has shown that the Ostionoid
phase of occupation is dominated by species preferring seagrass habitats, while the
Meillacan occupation was characterized by mangrove species of molluscs.
It is possible to think in terms of shell technology on this site, in addition to
the use of shell as a food source, due to the obvious and visible patterns of wear on
many specimens used as tools. It seems logical that early Redware populations on
Jamaica would have maximized their exploitation of an abundant shell resource,
which was not yet depleted by over-harvesting.' All species of shell found on the
site could have been collected easily by hand or with simple digging tools without
requiring complex tools or specialized fishing techniques. It is assumed that the
shell technology is oriented toward tasks or utilitarian work done in the daily lives
of these people within their settlement contexts.
Lastly, certain elements of the ceramic assemblage already show affinities in
form with other ceramic assemblages in the Greater Antilles, particularly with
some early forms on Puerto Rico (see for e.g. Espenshade, 2000). This situation
holds promise for establishing continuity between Jamaica's Redware peoples and
other Redware populations in the Greater Antilles. No doubt, greater numbers of
ceramic and other cultural correlations can be made in time, which would then
allow for a realistic assessment of the 'origins' of the Redware culture on Jamaica.
It is in fact more accurate to speak of the 'arrival' of Ostionoid culture on Jamaica
rather than an 'origin', since it is known by now that the Ostionoid phenomenon is
a distillation of at least two broad cultural entities (Central American and
Amazonian) that coalesced in the Greater Antilles c. A.D. 600. During the process
of Ostionan cultural evolution several island varieties of the complex were created
from regional transformations over hundreds of years until the evolution of the
Meillacan (Taino) sequences. Thus, prehistory in Jamaica is entirely a Greater
Antillean or northern Caribbean development restricted to the evolution of
prehistoric cultures within this region of the Caribbean. In time we may be able to
define all or most of the cultural affinities with Ostionoid populations of the
Greater Antilles, however, an additional challenge archaeologically will be in

determining the unique indigenous aspects of the Jamaican Ostionoid that are
different from its Greater Antillean counterparts.

1. I am pleased to acknowledge suggestions given by Thera Edwards for improving
the distribution map (Figure 2), and the assistance of James Robertson and Maaike
Lesparre-De Waal for their reviews of this paper. Versions of the paper were presented at
the Archaeological Society of Jamaica's sixth symposium on April 10, 2008, and at a
Department of History and Archaeology faculty/graduate seminar at UWI in Jamaica, on
April 18, 2008. Funding for the Blue Marlin Archaeological Project has been granted by
the Research Committee of UWI, Mona, to whom I extend sincere thanks.
2. I acknowledge the use of 'Google Earth' for the production of the satellite image
seen in Figure lb. All other photographs, maps, and illustrations in the article are by the
3. The culture is named for the type site of Ostiones in Puerto Rico. The nomenclature
is part of Irving Rouse's chronology, developed for the prehistoric sequences of all the
Caribbean islands (Rouse, 1992).
4. From Vanderwal, 1968 the complete calibrated date is A.D. 650 + 120, obtained
from the Little River site in St. Ann (see also table 1 of this article); from Keegan,
2000a, no calibration was reported for the date of A.D. 850.
5. New radiocarbon dates will soon be available from the Blue Marlin site, as
samples have been sent to the Beta Analytic Radiocarbon Dating laboratory, Miami, FL
for radiocarbon dating.
6. Multiple accounts of prehistoric migrations into the Caribbean region exist by now.
For some of these discussions see, e.g., Keegan 2000b, Wilson 2007, and Reid 2009.

7. The groups Pisces, Cheloniidae, and Aves require species differentiation by a
faunal specialist, a task soon to be addressed.
8. See Smith, 2000, chapter 3 for discussions of the seasonal behaviour of sea turtles
on and around the Cayman Islands in historic times.
9. See Smith, ibid., for colourful accounts of sea turtles caught in this manner in
historic times.
10. In historic times, turtle eggs were a "prized source of nourishment" (Smith, 2000,
p. 65).
11. The depletion of a number of endemic plant and animal species by prehistoric
populations is a well-documented fact for many of the Caribbean islands, including
Jamaica. See Wilson, 2007, p. 34 for a general discussion.
Atkinson, L.-G. (Ed.). (2006). The earliest inhabitants: The dynamics of the Jamaican
Taino. Kingston: University of the West Indies Press.

De Wolff, M. W. (1953). Excavations in Jamaica. American Antiquity, 18, (3), 230-238.

Ebanks, R. (n.d.). Paradise Park Amerindian (Arawak Indian) site at Paradise Park, Ferris
Cross, Savanna-la-Mar, parish of Westmoreland, Jamaica. W. I. A report of site survey
and test excavations carried out between February and June 1991. Unpublished raw data.
In the University of the West Indies Main Library, West Indies Collection, Kingston,
Espenshade, C. T. (2000). Reconstructing household vessel assemblages and site duration
at an early Ostionoid site from south-central Puerto Rico. Journal of Caribbean
Archaeology, 1, 1-22.
Keegan, W. (1998). Caribbean archaeology at the Florida Museum of Natural History
http //:www.flmnh.ufl. edu/caribarch/jamaicparpark.htm
Keegan, W. (2000a). Caribbean archaeology at the Florida Museum of Natural History:
Paradise Park, 2000. Retrieved September 1, 2008, from http //:www.flmnh.ufl.
Keegan, W. (2000b). West Indian archaeology. 3. Ceramic age. Journal of
Archaeological Research, 8, (2), 135-167.
Keegan, W., Portell, R. G., & Slapcinsky, J. (2003). Changes in invertebrate taxa at two
pre-Columbian sites in southwestern Jamaica, AD 800-1617. Journal ofArchaeological
Science, 32, (12), 1577-1617.

Lee, J. W. (1976). Jamaican redware. Archaeology Jamaica, 76, (2), 1-5.

Lee, J. W. (1979). The Jamaican redware culture. Archaeology Jamaica, 79, (1), 1-5.
Lee, J. W. (1980). Jamaican Redware. In S. M. Lewenstein (Ed.), Proceedings of the
Eighth International Congress for the Study of the Pre-Columbian Cultures of the Lesser
Antilles (pp. 597-609). Tempe, AZ: Arizona State University.
O'Day, S. J. & Keegan, W. (2001). Expedient shell tools from the northern West Indies.
Latin American Antiquity, 12, (3), 274-290.
Reid, B. A. (2009). Myths and Realities of Caribbean History. Tuscaloosa, Al.: the
University of Alabama Press
Rouse, 1. (1992). The Tainos: Rise and decline of the people who greeted Columbus.
New Haven, CT: Yale University Press.
Smith, R. C. (2000). The maritime heritage of the Cayman Islands. Gainesville, FL:
University Press of Florida.
Vanderwal, R. L. (1968). The prehistory ofJamaica: A ceramic study. Unpublished
master's thesis, University of Wisconsin-Milwaukee, Milwaukee, WI.
Wilson, S. M. (2007). The archaeology of the Caribbean. Cambridge: Cambridge
University Press.

' r
*V h^
- im


1 ...

I -,

Jamaican Taino 'Shellsmithing' Techniques
Explored: A Study in Method'


Jamaica was inhabited by two separate groups of pre-Columbian people.
The Ostionoid, or as James Lee called them, the Redware people arrived
approximately A.D 650 + 1202, settling several known sites on the southwestern
coast and on the central northern coast of the island as well. Then a later migration
of Meillacan Ostionoid pottery-making people presumably from central
Hispaniola arrived, approximately 1073 B.P3 + 95 or A.D 948 (Allsworth-Jones,
2008, p. 99). These were the Taino. They successfully settled most of the coastal
areas of the island as well as many inland sites.
Both cultures were technologically Ceramic Age peoples, who used the
durable materials around them to fabricate and develop their material world, the
hardwood trees, natural fibres, a large variety of stones, both igneous and
sedimentary, and the shells collected primarily as food, which they used
secondarily in their industrial and decorative arts. However, it is the indigenous
use of shells and corals which fascinated me. Having been a practicing goldsmith,
I understood the skill required in turning natural raw materials into naturally
fabricated things of beauty, but pre-Columbian peoples were far superior in skill
and ingenuity in their use of the seashell, compared to present-day manufacturers
of shell craft items. From a careful study of certain worked artefacts (presented
below) it appears that the Jamaican Tainos' understanding of the natural symmetry
and internal structure of various gastropod shells (Figure 1) was advanced and
equal to the skill ofMesoamerican (Mayan) craftsmanship (Schele & Miller, 1992,
plates 27, 59, 85, 113, 121), or Aztec, or indeed the Classic Taino craftsmanship as
seen on the other Greater Antillean islands of Hispaniola, Eastern Cuba and Puerto
Rico (Bercht et al., 1998, plates 25, 26, 34, 85, 102).
Though these cultures used Gastropod shells to make fantastic works of
wearable art, their artisans relied on incision techniques and carvings on the
surface of the shell to convey their ideas visually. The Jamaican Taino, however,
seemed to prefer the use of the natural internal structure of the shells' columellas to
translate and communicate ideas and imagery. Thus in my estimation their

understanding of the shells' true decorative ability was far superior to the other
previously named cultures.

Figure 1: Structure of a gastropod shell (Author's illustration).

va.ix or nowfle


Bol' wAlhorl

The most often used univalve shells were classified as the true conchs
(Dance, 1992, pp. 57-64). The term 'conch' is used to designate a number of shell
varieties, even those that do not belong to the true conch shell gastropod family as
do the Strombus gigas (Queen conch), the Strombus gallus (Rooster Tail conch),
Strombus raninus (Hawk Wing conch) and the Strombus pugilis (West Indian
Fighting conch), which are common Caribbean varieties of the sea snails. Living
in warm shallow waters, feeding on eel grasses and scavenging through coral
sands, they often move into deeper waters as they reach the end of their
reproductive maturity, although never to depths of more than 70 feet. At this depth
their shells, which consist primarily of calcium carbonate bonded by a protein
called conchiolin, start to thicken and lose their customary rippled appearance, as
in the case of the Queen conch (Strombus gigas). This rippling can be found on the
larger outer lip or flare of the Strombus gigas shell (Figure 2). However as this
shell ages the rippled surface starts to attain a flattened appearance. In some cases
this flattening is minimal and in others the flattening is complete, leaving a smooth
sloping surface. The snail can live for 20 to 40 years. This longevity also
influences the amount of horizontal rippling or ridges present on the body whorl,
which continues to the outermost rim of the lip as well as to the thickness of the
outer lip. The Melongena melongena or West Indian Crown conchs are also a
variety of sea snail, and are not considered true members of the conch family
though colloquially they are called by that name. They belong to the Melongena

Figure 2: Structure of a Gastropod Shell(underside) (Author's illustration).


Body ffowrl

family of molluscs, however, their life spans and diets are similar to the true conch.
Their shells as well as the shells of the true conch and several other mollusc
species4 are commonly encountered on the surface of Taino archaeological sites
across the island and were used in a vast assortment of ornamentation, as well as
in Taino utilitarian tool kits, both ceremonial and secular.

Techniques Used in the Manufacture of Shell Ornaments, Ceremonial
Objects and Tools
When admiring shell works of art one wonders what tools were used or how
long it might have taken to complete a single piece. The skill required was
remarkable considering the fact that the primary tool kit came from the same
materials as the ones that were employed in their manufacture. Often found on
archaeological sites across the island are grinding stones, polishing stones,
hammer stones, anvil stones, a variety of picks, plus an array of small micro and
the larger macro (chert or flint) blades. But there are other tools that were likely
employed by the Jamaican Taino which may not be easily located or readily
identified as tools either because they are made of vegetable fibres which are
perishable, or because of their nondescript appearance, such as wood or twine in
the first instance, and sand in the second. However simple these items may seem in
appearance their effectiveness in performing certain prescribed tasks as described
below invites discussion about the versatility of the Jamaican Taino technological
ingenuity. And in some cases these same techniques have remained unchanged in
the modem shell ornamental industry and goldsmithing in general (Untracht 1982,
The best way to explore the 'shellsmithing' technology of the Taino is to
examine specific artefacts and the techniques used in their fabrication. First on the

list of objects to be examined is a finely crafted shell pendant (Figure 3 a-c) found
at the Booby North Point Taino site (Figure 4), a site that was investigated and
named by the author on the 15th of January 1997.

Figure 3: An elegantly crafted pendant shown from three angles. North Booby Point,
Jamaica (Author's photograph).

Figure 4: Map of Eastern Jamaica showing the location of the Taino sites mentioned
above. Redrawn from S.A.G. Taylor, The Western Design: An account of Cromwell's
expedition to the Caribbean, map no. 1.

On recognizing that the artefact was pre-Columbian, the site was revisited,
and found to contain a further one hundred and forty-five shell, stone, and coral
artefacts, which were in peril of being washed out to sea. This ornament (Figure 3)
was constructed from a mature Hawks Wing conch shell (Strombus raninus),
which is more often found in the waters of Hispaniola. So it is not unreasonable to
assume that these Booby North Point Taino fished regularly near or in the Morant
Cays or as far away as Navassa Island5 (Figure 5), which is located only a few
miles from Hispaniola. This is still the practice of fishermen from the parishes of
Portland and St. Thomas today. Both parishes are located at the easternmost point
of the island.

Figure 5: Map of the island of Jamaica in relation to her neighbours. Redrawn and
redigitized from D.J.R. Walker, Columbus and the golden world of the Island Arawaks.
Map no.7, p. 235.

To make this object, after the extraction of the highly nutritious snail that
inhabited the shell, the pre-Columbian craftsperson first started his or her work by
burying the shell in an ant mound in a shaded location. This technique of ant

cleaning is commonly used by modem ornamental shell collectors to remove the
remaining fragments of the snail. Since harsh chemical treatments can damage the
value of the shell, ants are an efficient non-corrosive way of cleaning the shell. I
have even experimented with the use of ants, when preparing soup bones for use in
my own experimental jewellery, and other researchers have explored the use of
ants in the cleaning of skeletal remains (Crawford & Atkinson, 1975, p. 8). After
this process was completed to the artisan's satisfaction, he or she then commenced
with the conversion of the raw shell to wearable art by removing the spire (Figure
3a), either with a sharp and precise blow from a small hammer stone or with a
sharpened stone or shell chisel, such as the one shown in Figure 6. The shell was
held in place using a grooved anvil stone similar to the one that was found on the
same site (Figure 7) by one of my two field assistants, Ms. Claire Woods.

Figure 6: Damaged stone chisel found
at Chancery Hall, St. Andrew
(Author's photograph).

Figure 7: Anvil stone found at North
Booby Point (Author's photograph).

Next, the outer body whorl was removed, leaving the inner columella
encasing exposed. The columella is made up of several circular internal chambers
that increase in number as the snail increased in size, and it runs linearly from the
spire to the siphonal canal opening. In order to reveal its inner beauty, these
chambers must be individually exposed. The craftsperson would then use a small

pointed stone or chisel (Figure 6), likely derived from a hard crystalline igneous
rock i.e., quartzite or serpentine. This small implement was used to peck or groove
around the first outer section of the columella, thus creating a continuous incised
line or small stippled pits. These incisions helped to facilitate the removal of the
section with sharp precise blows from the small chisel when struck by a small
hammer stone. This would also have created less damage to the other underlying
columella segments of the shell (see Figure 2). They then moved on to repeat the
same process to expose the next section.
This was continued until all the required sections of the columella were
exposed, and then the Taino craftsperson inserted two small elliptical stones of
approximately the same size (as in Plate 3, COLOUR PLATES, p. v), each placed
on the opposite sides of the U-shape located at the centre of the other two
decorative openings. This well-chosen placement had the practical function of
helping to guide the suspension string or twine that was used to hang the shell
pendant when worn around the neck as a decorative ornament. However, there
would have been another purpose for the placement of these stones: I suspect they
represented a hidden symbolism when combined with the complete design as a
whole. This may have been associated with abstract thoughts and ideas as well as
representations of real images, both perceived and intangible.
There was likely a religious meaning connected with the choice of the shape
and design of the objects being made for ornamental or utilitarian purposes, as the
chief craftsperson of the village often would have been a member of the shamanic
caste and female. Taino religion was based on the existence of the spirit world that
was believed to have had constant interaction with the material world. Both
coexisted in harmony and sometimes disharmony; this meant that the vast majority
of ornaments were made to display one's affiliation with a certain protective spirit,
for example the owl. Several owl-shaped pendants were found at the Chancery
Hall Taino site (Figure 8). They all were made from murex shells simply altered to
look like the night raptor, a harbinger and protector of the dead (Arevalo, 1998,
pp.112-125). These facts were mentioned in ethno-historical accounts recorded
by Fray Ramon Pan6 (see Pan6, 1999; Keegan, 2007) and Las Casas (Ostaprowicz,
The earlier reference to the gender of these Taino artisans was verified by
local natives who told Bartholomew Columbus about a group of female
woodcarvers who lived on the island of La Gonave (Ostaprowicz, 1998, p. 66).
This island is located off the western coast of present-day Haiti as shown on the
map featured in figure 5. However, in pre-Columbian times the island was called

Figure 8: Murcx Shell, altered to depict the image of an owl (Author's photograph).

Guanaho" (Coll y Toste, 1972). This special guild of craftswomen apparently
made all the ceremonial trays, duhos 7 of polished black woods, basins and other
wood carved containers (Ostaprowicz, 1998, p. 65). These instances, however, do
not preclude Taino men from participating in the creative crafts, but many
historians and archaeologists have explored the dynamics of the matrilineal nature
of Taino society, noting how women controlled both lineage and the objects that
were associated with status and the caciques' authority to rule (see for e.g. Keegan,
1999, pp. 111-117).
The final step in making the shell pendant discussed above would be to
smoothen it with the use of a small J ln.AJiiLd ovoid shaped stone with a mildly
gritted surface texture, possibly selected from a nearby riverbed. A small stone
found at Chancery Hall, fits this description. It was approximately one inch long
and a millimetre wide. One side of the stone showed extensive flattening and
I-m.hllingi from its continuous use as a sanding tool. The stone would be
frequently dampened with water, to allow for an evenly distributed amount of
friction, while the piece was being sanded. The longer the stone was used, the
smoother and less gritty the surface would become, allowing the slow polishing of
the shell. This technique is still used today when sanding precious metals, but
instead of a stone, we use various grit sandpapers, :hlugll some jewellers will use
only one grade grit sandpaper with water to achieve the same outcome. As use
lessens the grit, it is possible to polish the jewellery piece to a fine matte finish.

Other techniques of shell fabrication were available to pre-Columbian
peoples on Jamaica as well. The second item to be examined would have required
more skilled tools than previously recognized on any archaeological site primarily
because of the perishable nature of the materials being used, but they were likely

employed by these Ceramic Age artisans in question. The tongue compressor
featured in figures 9a and 9b, is an artefact that was used by the Taino in their
cohoba8 purification ceremonies in a similar way in which the vomiting sticks of
Hispaniola and Puerto Rico were utilised (AllegriA, 1998, p.24 and cat. 14, p.28).
It was used by placing the broadest section of the compressor on the tongue and
thrusting it to the back of the throat. This caused the user to regurgitate or vomit
out any undigested food from the stomach. This was necessary before the Taino
religious practitioner could inhale the cohoba hallucinogen and enter a
transcendental trance-like state.

Figure 9 a, b, and c: A highly polished tongue compressor; provenance unknown
(Author's photograph and illustration).

Renwm al of oaer
drion ofste plare

This tongue compressor was cut from the outer lip or flare of a Strombus
gigas (Figure 9c) with a saw, not the type commonly known today but one that was
constructed with fibre twine tied to a small bow (Figure 10) or twine tied between
two hand sticks (Figure 11). Either variety would have been easy to manoeuvre
around comers and curves. The bow variety, however, was limited by its ability to
be easily manoeuvred unless the individual was properly seated. The bow saw
often requires the user to be seated upright and straight with the item placed
directly at eye level when being cut. This arrangement would have been highly
unlikely since the only known chair recorded as having been used among the
Taino was the caciques' duho (Ostaprowicz, 1998, pp. 56-67), which either had a
curved seat with a high back or was a low ornately decorated stool. On the other
hand, all the craftsperson needed in order to use the dual-stick variety (Figure 11)
was a good seat on the ground or a low stool, or a stone with some hardened clay to

fasten the shell in place. She (or he) could also use both feet as a vice-grip while
seated on the ground and holding the shell in place.

Figure 10: Example of a bow saw made from curved wood and twine (Author's drawing).

Figure 11: An example of the dual stick handsaw (Author's drawing).

/ - -- -

The fibre twine could have been obtained easily from any non-siliceous
plant, i.e., soft grasses, the Agave leaf or any cave-hanging liana vine. The woody
surface of these plants would be striped off, leaving the soft fibre, which was
scraped and stripped down further to the width and thickness that the artisan may
have required, then sun dried. At the Chancery Hall site (Figure 4) a juvenile conch
shell was discovered which had a long grooved notch chipped into the shell, with a
rounded posterior end (Figure 12). This showed signs of having been retouched on

the inside of the curvilinear notch, making it sharp enough for stripping
non-siliceous fibres.

Figure 12: A juvenile conch shell altered to strip twine. (Chancery Hall, St. Andrew,
Jamaica (Author's photograph).

It appears likely that the alterations to the shell shown in figure 12 were done
by a specific Pre-Columbian artisan primarily for the purpose of striping fibres to a
prescribed width. It can not be inferred, however, that this tool was definitely used
for making twine for a dual stick saw, but the technology of twine stripping is
strongly suggested by this tool's discovery and analysis. As is done today in
jewellery making, several different sized saw blades are used, from blades that
measure only one-eighth of a millimetre in thickness to several centimetres. Each
size serves a different purpose from cutting to rasping small edges. In the Taino
case, the fibre twine was attached to the bow or sticks. A vegetable resin was likely
applied to the twine and a layer of sand was rubbed over the resin-soaked twine
and left to harden again. Now it was ready to be employed in the making of the
tongue compressor. The shape desired was marked out on the shell flare with a
piece of charcoal, and now the slow deliberate cutting could begin. As the twine
lost its sandy grit, more fine-grained sand was drizzled onto the area being cut.
With this continuing action the blank tongue compressor could have been cut
within a few days of continuous work.
This technique was also used on the paddle-shaped artefact (Figure 13),
which I suggest could have been used by women when cooking and shaping
cassava bread on their clay cooking griddle. The 'blank' for this artefact is

illustrated in figure 14, as it would have appeared prior to its removal from the
Strombus gigas shell.

Figure 13: A paddle shaped artefact. Figure 14: Paddle shaped artefact
Martello Tower site, Harbour View, before it was cut from Queen conch
Jamaica (Author's photograph). shell (Author's illustration).

Other cutting methods were also utilized by the Taino, particularly when
cutting small objects, like the circular clamshell beads shown in figures 15a and
15b. These beads would have required a sharp pointed awl (Figure 16) in order to
make them. This example was found at the Chancery Hall Taino site,9 along with
the beads shown in figure 15. To extract the circular shape of the bead, the awl
would have been used to lightly engrave a circular line on the shell surface, until
the groove grew deeper and deeper, then the artist would lightly tap away the
remaining shell which surrounded the deeply engraved circle using a small chisel
or hammer stone.
The same awl, when turned on its side, could have been used to sand the
rough edges off the blank bead, because in this case, the awl seemed to have come
from a gritty silicon-based stone. It has a rough enough surface to be used in small
sanding work, and it is also quite similar in shape to small rasping tools used by
modern jewellers to smooth and round the edges of disks. With a small amount of
fine-grained sand the awl was used to drill the centre hole by holding the bead on a
fingertip used as support for the convex side of the shell and rotating the awl back
and forth on the bead's concave side. This process by my reckoning would have
been the most effective way to make the beads in question, since a similar method

of polishing small hand-held jewellery items is used universally by jewellers
today, even though when we drill small objects we use an electric drill quite
similar in type to a dentist's flexible shaft drill. When drilling small domed objects,
the modem jeweller will place them on a wooden surface that has a pre-cut notch
to support the dome and they are best drilled from the same direction as in the
pre-Columbian case. However, this method is not strictly adhered to by modern
jewellers. On realizing that modem tools would not have been available to my
pre-Columbian forebears, I imagined the process using a prehistoric awl, which I
then experimentally replicated with a sharpened nail and discovered that though
not identical in result, it was possible to make a bead in this manner.

Figures 15a and b: Shell beads from Figurel6: A quartzite awl from
Chancery Hall, St. Andrew, Jamaica Chancery Hall, St. Andrew, Jamaica
(Author's photograph). (Author's photograph).

The construction of the handled snuff holder/tray featured in figures 17a and
17b was fabricated using a very simple procedure. This artefact has been labelled
as a snuff holder/tray by me because of its strong visual similarity to more
elabourately carved wooden ethnographic examples which have been both studied
and observed in use in the Amazon delta (McEwan 2001, pp. 192-195), a region
whose cultural practises are not very dissimilar to those of the Jamaican Taino.
The artefact was made after all other parts of the Queen conch (Strombus gigas)
had been removed for the creation of other ceremonial or utilitarian items (Figure
18). What would then be left is the columella and the bottom half of the body whorl
and inner lip. Since in this instance the columella was not needed it too was

removed, leaving behind a slightly curved triangular-shaped dish, which used the
siphonal canal, inner lip and lower aperture for the tapered handle.

Figures 17a and 17b: A large conch shell handled cohoba snuff holder/tray. North
Booby Point, Portland, Jamaica (Author's photograph).

Figure 18: Snuff holder/tray before it has been removed from the shell (Author's

A Brief Examination of Tool-wear Markings on the Booby North Point
In order to offer a full analysis of the methods used by the pre-Columbian
Jamaicans, we can examine a single artefact in detail to understand their
technological techniques. This is, however, not a suggestion that these same
methods might not have survived into the post-Columbian era. But for this
examination, the elegantly crafted pre-Columbian pendant found at Booby North
Point featured above in figures 3a to 3c was chosen. This artefact was magnified to
5x magnification, revealing very distinctive traces of tool wear markings, which
supported the previously discussed theories of its manufacture. These will now be
described briefly through the use of figures 19-21.
Around the vicinity of the removed apex (Figure 19), the rounded rim can be
observed clearly in this close up of the area. On the columella entrance a notched
or stepped feature was created to recess the area. This also shows rounded
smoothing, indicative of the artisan's use of a small sanding tool.

Figure 19: Close-up of the removed spire opening (Author's photograph).

The uppermost point of the apex was removed, (Figure 3b above) revealing
the spire belt which shows evidence of having been polished in a clockwise
motion, leaving it perfectly rounded along the entire belt's contour and edges.
Three grooves were filed or notched, which included a centrally located groove
that was separated by a small bridge-like area of unremoved shell material from
the exposed coil-shaped area on the exposed upper inner columella. The two outer
notches on either side of the spire belt are the continuing comers of the uppermost

U-shaped decorative feature located on the columella, shown partially in Plate 3,
(p.v,) which is described below.
The two oblong stones (COLOUR PLATE 3, p. v) were placed securely in
the natural curvature of the shell and were likely hammered lightly into place,
however there are a few small unaltered chips on the edge of the shell beside the
stone on the left side, which may be interpreted as evidence of remaining hammer
marks. This suggests that the sanding of the shell occurred subsequent to the
stones' placement.
The frontal U-shaped areas both show an inward curving which resulted
from the shell being sanded in a continuous upward direction (starting from the
base, i.e., the siphonal canal opening towards the removed apex). The
unidirectional sanding technique is still used today by goldsmiths, primarily to
create a smooth polished matte surface. However, it also can be noticed that in the
same location on the shell many small chips are observable. These are the result of
the chipping method used to remove that section of the columella.
At the back of the shell (Figure 20) several other features are observed,
namely translucent thinning near the left hand corer of the horizontal U-shaped
opening. This is a result of aggressive sanding or polishing methods. A chip is also
present, possibly caused by the area's loss of material stability, leading to a piece
of the shell falling off during sanding. This loss of stability is also evident in the

Figure 20: Close-up of the decorative feature at the back of the shell pendant (Author's

formation of faint radial fractures, which can be seen just below the outermost
curve of the shell's body whorl nearest to the right side of the horizontal U-shape.
However these small imperfections were not major enough to cause any dramatic
damage or to change the overall durability of the shell pendant.
Finally the base of the shell pendant (Figure 21) shows some of the most
conclusive evidence of the skill and competence exhibited by the shellsmiths of
Booby North Point. Here, at the base, the shell has perfect symmetry and the
polishing is flawless.

Figure 21: Superior polishing skills of the artisan, seen near the siphonal canal opening at
the back (Author's photograph).


This is a brief overview of a study which is still ongoing, and which should
lead to a fuller comprehension ofpre-Columbian shell-working techniques and the
tools used in Jamaica's pre-Columbian societies. Much of the shell from Jamaican
Taino sites has been counted and categorized; the time has come for them to be
studied in conjunction with, and more specifically, alongside all areas of
archaeological and ethnographical research, as has been done for ceramics.

Although other attempts have been made to identify various shell tools and their
uses (e.g. Alvarez, 2007, pp. 226-242) or even how some were made, much more
needs to be done here in the Caribbean. The interpretive work on the Craig Mound
shell artefact cache found at Spiro, Oklahoma in the southeastern United States
(Phillips & Brown, 1978), which was looted by commercial artefact profiteers in
the 1930s, offers suggestive comparisons for the Caribbean region. At that site, a
comprehensive attempt was made not only to collect rubbings and drawings of the
fragmented and scattered artefacts, but also to classify the various styles of
artefacts. This resulted in the identification of the possible existence of workshops,
schools and individual artists with their artistic peculiarities. This research has led
to a deeper understanding of the artistic dynamic behind the Southern Cult
complex, which existed among the Mississippian culture of the southern United
States, colloquially called the Mound Builders.

Thus I am suggesting that a multi-disciplinary approach like the one used at
the Craig Mound site be applied here in the Caribbean. There should also be much
more consistent and comprehensive attention in reports to potential information on
the techniques of manufacturing, and to ornamental use patterns, as well as to how
these patterns may have changed over time, along with a study of iconographic
similarities or dissimilarities between various cultures and tribal groups in Jamaica
and the wider Caribbean Diaspora. The questions that must be asked now are:

Did the Taino have specific workshops for the production of the shell
artefacts and utensils?
Were they located on a particular location on a site?
Was this art form practiced exclusively by a separate class of artisans?
How were the ornaments worn?
How were these ornaments divided among the population, in other
words which items were worn by which socio-political group or social
As my final comment I believe my paper is only the start of a wonderful
adventure in the study of prehistoric shellsmithing.


1. It is my pleasure to acknowledge Sharon Tulloch for providing regular
transportation to the Chancery Hall Taino site, and Claire Woods for helping me explore
both the Chancery Hall site and the Booby North Point site. I especially thank Diane
Crooks for showing me the shell artefact shown in figure 3, and for her willingness to
show me where it came from; this led to the discovery of the hitherto unknown Booby

North Point site. I would also like to thank Mr. Ivor Connolly of the Dept. of History and
Archaeology at the UWI (Mona) for providing me with specific information that has
helped me to classify and catalogue the artefacts more efficiently. And finally I wish to
thank both Dr. Sabrina Rampersad and Dr. James Robertson for their reviews of this
2. The date referred to is the recognized date for the arrival of the Ostionoid ceramic
peoples locally called the Redware people, based on radiocarbon dates derived from
Little River, St Ann and Alligator Pond (Bottom Bay), Manchester (see the discussion in
Atkinson, 2006, p. 3).
3. This date is based on carbon dating samples taken from the White Marl (Meillacan)
4. Some of the other varieties of shells which were used by the Jamaican Taino to
make ornaments and tools were the West Indian Murex (Murex brevifrons), Bent-Beak
Murex (Murex recurvirostris), Trumpet Triton (Charonia variegata), Angular Triton
(Cymatium femorale), Atlantic Partridge Tun, the Flame Helmet (Cassis flammea), the
King Helmet (Cassis tuberosa), the Netted Olive (Oliva reticularis), Caribbean Olive
(Oliva scripta, the Mouse Cone (Conus mus) and the West Indian Top shell (Cittarium
pica), to name just a few. The bivalve shells used were a variety of Oysters, Cockles,
Venus's, Clams, Tellins, Strigillas and members of the Glycymeris family (see Sutty,
5. Navassa Island was called Nabaca by the Hispaniola Taino.
6. Las Casas noted that the island located off the western coast of Haiti, was
subordinate to the Caciquedom of Xaragua. This Caciquedom was ruled by Behechio and
later by his sister Anacaona, the wife of Caonabo. The island was called Guanabo by the
Indians. In a dictionary of indigenous words compiled by Dr. Cayetano Coll y Toste, a
pre-eminent Puerto Rican historian, he noted that in modern times this word was
corrupted to Gonaive. Gonaive, however, was a likely attempt to convert a Taino word
into a word more compatible with the syntax of the French language, which was
subsequently spoken in Haiti after the island was split by the two rival powers of France
and Spain.
7. Ceremonial throne-like seats used by a Taino cacique, they are made primarily
from hard woods, i.e., lignum vitae or mahogany.
8. Cohoba was made from the seeds of the Anadenanthera peregrine or Piptadenia
peregrine trees. The seeds were ground with crushed shell or sometimes tobacco to
enhance its hallucinogenic properties.
9. Specifically found on Lot no. 380, located on Lord Nelson Drive, Kingston,

Allegrid R.E. (1998) An introduction to Taino culture and history. New York, Monacelli

Allsworth-Jones, P. (2008). Pre-Columbian Jamaica. Tuscaloosa, AL: University of
Alabama Press.
Alvarez, A. R. (2007). The Archaic shell tools ofPunta Las Cabezas de San Juan, Puerto
Rico. In Reid, B., H. P. Roget, and A. Curet (Eds.), Vol. 1, Proceedings of the
twenty-first congress of the International Association for Caribbean Archaeology (pp.
226-242). St. Augustine, Trinidad: University of the West Indies, School of Continuing
Ar6valo, M. A. Garcia (1998). The bat and the owl: Nocturnal images of death. New
York: Monacelli Press.
Atkinson, L.-G. (Ed.). (2006). The Earliest Inhabitants: The Dynamics of the Jamaican
Taino. Kingston, Jamaica: University of the West Indies Press.
Brecht, F., Brodsky, E., Farmer, J. A., & D. Taylor, D. (Eds.). (1998).Taino :
Pre-Columbian art and culture from the Caribbean, New York: Monacelli Press.

Coil Y. Toste, C. (1972). Vocabulario indo-antillano, Cldsicos de Puerto Rico. 2nd ed
Accessed from:http:// members.dandy.net/-orocobix/termsl.htm in January 2001
Crawford, R. L. & Atkinson, J. B. (1975). Fire ants used in skeletal preparations. The
Florida Entomologist, 58, (1), 8.

Dance, S. P. (1992). Shells. London: Dorling Kindersley.
Jones, S. & Jones, A. (2005). Seashore life of the Caribbean. Basingstoke: Macmillan
Caribbean Press.
Keegan, W. F. (1999)."No man (or woman) is an island": Elements of Taino social
organization.Gainesville, Fl. University Press of Florida.
-. (2007). Taino Indian myth and practice: The arrival of the stranger king.
Gainesville, FL.University Press of Florida.
McEwan, C. (2001). Seat of power: Axiality and access to invisible worlds. In C.
McEwan, C. Barreto, and E. Neves (Eds.), Unknown Amazon: Culture in nature in
ancient Brazil. London: British Museum Press.
McEwan, C., Baretto, C., & Neves, E. (Eds.). (2001). Unknown Amazon: Culture in
nature in ancient Brazil. London: British Museum Press.
Ostapkowicz, J. M. (1998). To be seated with 'great courtesy and veneration':
Contextual aspects of the Taino duho. New York: Monacelli Press.
Pan6, R. (1999). An account of the antiquities of the Indians, J. J. Arrom & S. C.
Griswold (Eds.). Durham, NC: Duke University Press.
Phillips, P. & Brown, J. A. (1978). Pre-Columbian shell engravings from the Craig
Mound at Spiro, Oklahoma. Cambridge, MA: Harvard Press.


Schele, L. & Miller, M. E. (1992). The blood of kings: Dynasty and ritual in Maya art.
London: Thames and Hudson.
Sutty, L. (1990). Seashells of the Caribbean. London: Macmillan Educational Ltd.
Taylor, S. A. G. (1965). The Western Design. An account of Cromwell's expedition to
the Caribbean. Kingston: Jamaica Historical Society and the Institute of Jamaica.
Untracht, 0. (1982). Jewelry: Concepts and technology. London: Doubleday.
Walker, D. J. R. (1992). Columbus and the golden world of the Island Arawaks.
Kingston: lan Randle.

Archaeological Investigation of Suriname
Maroon Ancestral Communities


The following report outlines archaeological findings from two
eighteenth-century Maroon ancestral settlements in Suriname, South America.
Maroons are descendants of escaped slaves and their social development presents
a unique perspective of African-Diaspora material culture. The initial objective of
the archaeological research was to identify, survey and excavate ancestral
settlements for artifacts depicting kin-group divisions and clustering and material
culture resources.
In August and September of 1997, members of the Maroon Heritage
Research Project (MHRP) identified the Saramaka site, Kumako and Matawai
site, Tuido located in the Suriname and Saramaka River basins, respectively
(Figure 1).
The location and general historical overview of each ancestral settlement
was determined from the available body of literature on Suriname Maroons, more
specifically, the Saramaka (Bos, 1998; Fermin, 1781; Goslinga, 1971, 1979, 1985;
Green, 1974; Hoogbergen, 1990; Jackson, 1965; Khudabux, 1999; Klooster,
1997; Price, 1975, 1983a, 1983b, 1990, 1996; Price and Price, 1980, 1988, 1991,
1992, 2003; Stedman, 1796). The project progressed to collecting ethnohistorical
accounts on the place names, and building liaisons with various Maroon governing
bodies (notably, the then Saramaka paramount chief and the current Matawai
paramount chief). All these avenues of inquiry generated questions about the
settlements for village elders and key individuals in the Maroon social structure,
who would later become integral to the overall success of the project. The field
excursions were successful in that they led to the identification of both these
settlements (see Figure 1). The discussion that follows describes the objectives
designed for each archaeological site and subsequent excavation season.

Figure 1: Research region depicting Maroon ancestral settlements and the current scope
of archeological research


S/Mountain Range
SHistorical Settlement
Archaeological Site
Contemporary Settlement

0 10 20 30 40 50

Saramaka Site, Kumako
During the 1997 survey season, members of the MHRP crew were escorted
to Kumako by a village elder who are the individuals most knowledgeable about
the history of the site. The initial travel route and destination were pre-plotted and
tracked with a hand-held GPS device and then mapped. Upon arrival at the general
coordinates (4 24' 50.0394" N latitude and 55 27' 43.6206" W longitude), a rock
structure was identified that still remains as a marker for the northernmost
boundary of the site. The site sits at a mean elevation of 293 metres above sea
level.' Areas with surface artifact scatters were identified and recorded with a
hand-held GPS device. Artifacts included a musket ball cache and fragments of
ceramic pottery. The discovery of these items further substantiated the location of
the site, but no excavation took place at this time.
Excavation of the Kumako settlement began during the 2000 fall field
season. Multiple goals were considered at this time. The first objective was to
create a path through the forest to connect the Suriname River Valley with the
Saramaka River Valley, passing either north or south of the Ebba Top Ridge. This
path would allow for easier interior forest travel and ultimately foster
simultaneous excavation at multiple sites. In addition, it would aid in the
understanding of how early Saramakaans and Matawais may have used these same
methods for strategic travel and communication between the major rivers.
Although MHRP members were unable to create the forest path, some helpful and
important information was gathered. First, it gave us a sense of the terrain, flora,
fauna, and conditions of the interior forest. This area is no longer inhabited by
humans due to its distance from major waterways. Secondly, some of the crew
who made it to the Ebba Top peak were able to take GPS readings along the way.
These GPS points allowed us to get a clearer idea of where Ebba Top was situated
with regard to Kumako. Thirdly, along the way, both crews came across other
potential sites-flat areas near water that appeared to have been cleared earlier in
the history of the area. Moreover, we realized that clearing a path connecting the
two river valleys was too impractical and labour intensive given the time frame
with mainly student volunteers, forcing us to focus on excavation at one river
valley per field season (personal communication from Becker, 2000).
Excavation continued at Kumako, where two distinctly anomalous sections
of the settlement were given the broader demarcation title of'Area'. Smaller zones
of interest within an Area were designated as 'Loci'. In the fall of 2000 Area A,
Locus 1 and Area A, Locus 2 were established (the details of which will be
discussed below in the section titled KMK1). The 2002 excavation saw a fairly

large amount of archaeological work accomplished. A part of the site
(approximately 8 m x 10 m) adjacent to the field crew's camping ground was
excavated. The artifacts (mostly worn or eroded ceramics) recovered at this time
produced a large majority of the total artifact yield from the Kumako site.
The 2004 excavation season at the Kumako site was distinguished from
preceding excavation seasons by the discovery of a large circular mound. At this
time our field crew included a noted hunter from Tutubuka. The hunter suggested
that we extend our survey to another nearby expanse of land loosely referred to as
the village proper or true village of Kumako. The survey led to the identification
and excavation of the mound site named Kumako 2 (hereafter referred to as
KMK2). While the initial Kumako was observed as a place of relative openness
vis-a-vis the surrounding forest, it contrasted starkly to the circular structure that
demarcated KMK2, a mound that went unnoticed (because of its subtle slope) by
crew members during the excavation seasons of previous years. KMK2 was
defined and excavated during the excavation season extending from August 2004
through January 2005.2 The earlier excavations of the Kumako settlement will
hereafter be referenced as KMKl.
Matawai Site, Tuido
The 1997 survey season also witnessed the location of the Matawai site,
Tuido. Tuido's importance stems from its role in recent Matawai history. As a
settlement contemporaneous with Kumako (established in the early 1700s), Tuido
was unique because it remained in use among contemporary Matawai until the
early 1900s. The settlement was visited by Dr. Kofi Agorsah following the
direction of a Matawai elder born there.3 Owing to the distance from easy travel
routes, as well as lack of roads and other communication systems, the trip to Tuido
was made from the capital, Paramaribo, by chartered plane to Pusugrunoe
(Poesoegroenoe). Travel continued south with an outboard motor-propelled canoe
for a two-day voyage on the Saramaka River to Tuido. Tuido was identified by the
same methods as Kumako (by using pre-established GPS coordinates for earlier
descriptions and known creeks) and has an intentionally clandestine location
along the Tukumutu (Toekoemoetoe) Creek in the Saramaka River Valley (see
Figure 1, above).
The site is located in a bend on the Pikein Tukumutu Creek, approximately
six kilometres westward from where it diverges from Tupi Creek. The location
corroborates well with Matawai and Saramaka oral traditions, which place the
settlement on the Pikein Tukumutu Creek (Price, 1983a). Tuido purportedly
consisted of many different settled groups or sections. In addition, Tuido is just a

few miles down the Tukumutu Creek south of Djomasanga (an early settlement
that has been confirmed by oral traditions and Maroon guides). Tuido's general
coordinates are 4 4' 59.9982" N latitude and 56 0' 0" W longitude.
Locating both the Kumako and Tuido settlements established the agenda for
subsequent archaeological research. Although Tuido was identified and its
boundaries established, it was not excavated-only surface collected. Our focus
was instead turned to the logistically more accommodating site, Kumako. The
findings from Kumako created an even greater impetus to return to Suriname with
a research design to accommodate the wider context for these major Maroon sites.
Archaeological Investigation: Sampling and Excavation Methods
The bulk of the archaeological evidence under discussion here is derived
from results from the 1997 through 2004-2005 excavation seasons.
Archaeological expeditions will hereafter be addressed according to the year of
the excavation.


1997 Surface Survey
1997 Surface Survey
1998, Rock Structure

2000, Area A, Locus 2
2000, Area A, Locus 1
2002, Camp Site

2004-2005, Kumako Mound
During the research years, we relied upon four reliable Maroon guides, who
were familiar with creating and tracking paths in the forest, as well as the overall
objectives of MHRP (at least one of these individuals has been present for each
excavation season since the project's inception in Suriname). Another three to four
Maroon men aided with the general tasks of camp setup and excavation. One
English-speaking Maroon colleague/research assistant functioned as a translator
and logistical coordinator for all tasks relevant to the organization and completion
of each excavation season. All work at the site, including survey and excavation
was carried out in ten-day rotations, in addition to two-day treks (to and from the
village base camp of Tutubuka), and a five-day break during which artifacts were

cleaned and catalogued. Digging procedures were also reassessed, and supplies
restocked from Paramaribo.
The entire Kumako site was mapped with a GPS, and an excavation grid was
applied to maintain spatial control of surface and subsurface artifact distributions,
as well as the position of features, such as circular depressions, knolls, and soil
stains (McMillon, 1991). All units were excavated in 10 cm intervals. In the earlier
expeditions, soil samples were collected from identified features to assess the
site's geochemical composition. We intended to test the soil for phosphorus (P)
concentrations. These tests would help to assess the chemical signature of soil that
may suggest specific lived behaviour, such as food production areas or refuse
deposits (Dillon, 1993; McMillon, 1991). We believed that testing for phosphorus
(P) would aid in substantiating ethnographic information about exploited natural
resources, subsistence, dwellings, and artifact manufacture. However, the caliber
of the samples did not allow for an isolated and therefore accurate reading.
Different screen sizes were used to collect various types of remains. A
1/16-inch mesh was used on identifiable features to collect macroscopic organic
remains and smaller artifacts. A 1/8-inch mesh was used for general soil sifting to
collect larger artifacts. Upon completion of excavation, the site was backfilled and
left to natural regeneration. All materials have been archived at the Suriname
National Museum in Paramaribo.


The initial survey of Tuido was conducted during the 1997 surface survey.
The site was divided into four areas demarcating sections marked by mound
clusters and differing concentrations of surface assemblages were identified. Each
area with mound clusters is also marked by groups of large trees and open land
areas. The mounds are believed to represent collapsed huts or rubbish dumps.
Other surface features included clay hearths, stone circles, and approximately 70
large and small pieces of black to dark brown earthenware-including large flat
pieces, rim and body fragments-and imported stoneware. Several areas were
more productive although it is premature to make any conclusions regarding the
artifact distribution at the site based on a limited surface study. Because this site
was not excavated, but identified, surveyed, and recorded, surface finds embedded
in the soil were removed.

Tuido is a very dynamic settlement and will require greater archaeological
attention in the form of surveys and excavations before we can engage in a full
dialogue. Nevertheless it deserves mention and consideration because it
contributes to the broader discussion of Maroon materiality.

Figure 2: Map of Kumako settlement, KMK1 vis-a-vis KMK2

_j ., De re Old Growth Forst A

Acomgandi Creek

or I ... OfI




+ Excavaions
0 KMK2 Mound Otch
KK2 Mound 0 375 75 150 225 300
KM2 Mound Roads 2Meter Sope Meters
A Datum Temporary Bench Mark (Tr

Kumako (KMK1)
The initial survey of KMK1 and its subsequent excavation focused on what
was initially thought to be the entire settlement. The major attributes identified
were located during 2002, adjacent to the crew's campsite. The crew's campsite
was also part of the early settlement. The campsite fits an earlier description of
relative openness vis-a-vis the surrounding forest. Prior to 2002, a survey of the

campsite was conducted. In this setting such tasks as maintaining consistent linear
transects with a compass, proved to be a challenging undertaking. Much of the
surveying consisted of walking tight 5 m transects in open and forested areas in the
vicinity of what historical and ethnographical evidence both described as
Kumako. In addition, an excavation grid consisting of 2 m x 2 m units was applied
to the region adjacent to the campsite, where a concentration of surface artifacts
was identified. Artifacts were removed as surface collections during the transect
survey. Once excavation began, the artifacts were recovered in 10 cm arbitrary
intervals and the entire matrix was sifted. The KMK1 site, as originally
demarcated, was divided into four areas, each about 200 metres square, totalling
approximately 16 hectares and forming an oblong circle defined by its relative
openness and the appearance of surface artifacts (Figure 2).
Both the surface collections in this open region and the excavations
produced more than 800 ceramic sherds to a maximum depth of 40 cm below
surface. Due to erosion and abrasion 92% were identified as rim sherds, but with
no identifiable attributes. The remaining 8% consisted of rim potsherds but with
no identifiable attributes.
In 1998, an area near a large rock structure was excavated. The rock
structure was identified during survey because of its proximity to a water source
and the relative openness of the surrounding forest. One 1 m x 1 m shovel test pit
was placed near the rock. A yield of 39 potsherds was recovered from the first ten
centimetres below the surface, beneath which sterile soil was quickly reached.
Based on these counts of ceramic sherds, the area near the rock structure, the
northernmost region of Kumako, was not excavated further. In addition, the dense
vegetative area extending beyond the immediate region of the rock structure
offered little opportunity for surveying and showed no surface scatters of artifacts.
This large rock is important because it is a fixed structure on the landscape and at
the site. More importantly, the rock structure demarcates the point of access from
the footpath from the river village, Tutubuka. These findings represent the
MHRP's early attempts at identifying the borders of the Kumako settlement and at
understanding the general distribution of where and how early Saramakaans
occupied the region.
During the 2000 season Area A, Loci 1 and 2 of KMK1, were tested. At this
time the earthenware and glass found at KMK1 were uncovered. In Area A, Loci 1
and 2 of the site, we recovered a total of 211 ceramic fragments, 88% identified as
potsherd rims. However, during the analysis process the rim type could not be
determined (these artifacts were however analyzed for temper). Area B is defined

by a cache of musket balls, but was not excavated due to time constraints and a
desire to explore what seemed to be more promising parts of KMKI.
The most important excavation at KMK1 took place in 2002 at the campsite.
This area is important because it held an array of artifact types (glass, metal, and
earthenware) that may help in understanding different Maroon activities. Figure 3
represents the other excavation units at KMK1. The crew's campground is
separated from the excavation units by dense vegetation and the camp lies at the
western boundary of the site. The campsite was designated after a noticeable
surface scatter was identified in an open area, free of dense ground vegetation. An
excavation grid was plotted and several units were opened and excavated to the
level of sterile soil, which occurred at 40 cm below surface.

Figure 3: Campsite excavation (Photo by MHRP Member).

The campsite units exhibited an array of artifacts similar to those found in
Area A, Loci 1 and 2. The cluster of units that define the campsite and the artifacts
occurring there are isolated and found in this part of Kumako only. The artifacts
removed from the excavation units adjacent to the crew's campground were
mainly worn and eroded ceramic sherds. Although this part of the site also
produced metal and high yields of ceramics, the difference is the appearance of
sherds of dark green glass isolated in this central region of KMK1. What we have
recovered from KMK1 to date is but a slight representation of the material nature

of the Kumako settlement. The 2004 excavation season of KMK2 presented
definitive features and structures in the form of a mound (which I will discuss at
length in the next section) and some significant traits in contrast to KMKI.
Kumako (KMK2)

The 2004 excavation season was distinguished by the discovery of a large
circular mound. The mound was discovered in August 2004 with direct guidance
from a Saramaka hunter familiar with the forest terrain. The entire excavation
season was dedicated to understanding the depth and general characteristics of the
mound (Figure 4). The boundary of the mound was determined with a GPS outline
of what was initially identified as a ditch-like feature. GPS readings defined the
perimeter of the mound. The mound is approximately I kilometre in
circumference and bordered by the ditch, which sits slightly below ground level
vis-a-vis the surrounding forest. Fallen trees and an overgrowth of old vegetation
pose some difficulty in accurately determining a consistent depth of the ditch and
therefore depth was not measured. The ditch ranges in width from 1 to 2 metres at
various points around the mound. The points where the mound periphery and the
ditch meet at ground level are intersected by roads or paths leading into the mound.
Four such roads were identified and recorded. The southern side of the mound is
characterized by sharply sloped edges, approximately 2.5 to 3 metres from the top
edge to what is identified as the ditch floor.
The construction appears to have created a defensive barrier that allowed
maximum strategic and visible advantage for those on the mound. Wooden
palisades were observed circling Maroon villages during militia battles against
Maroons in the early 1730s (Price, 1983a), an observation that fits with these
configurations, but we have not yet documented archaeological remains for a
palisade (Figure 5).

The ditch encircles the mound and is bordered by old growth trees that
continue into the outlying forest. The ditch is easily identified in contrast to the
edge of the mound at its northeastern periphery. As obvious as the ditch appears at
the northeastern periphery, it is less conspicuous at various points around the
The northeastern edge of the mound is in alignment with an access path
created by the crew. This was not an obvious path but instead was distinguished by
small chops in trees and/or broken branches to mark the way. This is the standard
method used by hunters to create paths in the dense forest. This path extends
approximately 70 metres connecting KMK1 (the crew's campground) to KMK2.

Figure 4: Map of KMK2



+ Ejv
O KWOauon mldn
a* Ka bund
/ AtrpdCmxk

60 30 0 06
S2 Meter Slope Melos

The path and the northeastern periphery of the mound mark a T-section that forces
one to ascend the mound, travel into the mound via Road 1 (see Figure 4), or
follow the ditch either east or west, thereby encircling the mound.

Upon ascending the mound, a large exposed area was immediately
identified at the mound's northeastern periphery by the MHRP crew. A 24 m x 12
m excavation grid was created to further investigate the large exposed area
(COLOUR PLATE 4, p .v). The exposed area was characterized by a dark circular
depression approximately 60 cm below the surrounding ground surface from its

centre dip, roughly 3 metres in diameter and, as we later learned during
excavation, 2 metres deep. The excavation units demarcating the dark circular
depression includes the 2 m x 2 m units H22-24 and 122-24 (see COLOUR
PLATE 4, p.v).

Figure 5: A schematic drawing of the mound edge and ditch floor

KMK2 Mound Surface -2.5 meters


Co impact old vegetation

The dark circular depression was described by the Saramaka field crew as a
product of ritual baths used by elders. It produced a large yield of artifacts ranging
in type: non-diagnostic grit and sand-tempered earthenware; red-slipped ceramics,
some with a patina film; and one piece of burnt Delftware.
Unit H22 from the periphery of the excavation grid was the most striking
due to a maximum depth of 1.50 m and a dense concentration of artifacts.
Excavation began at the centre of the depression, first removing the loose topsoil
layer of reddish brcwn clay loam with heavy root disturbance. As excavation
continued and the dark circular depression borders diffused, the density of artifacts
shifted toward the northwestern part of H22 and a Im x 1 m unit was started at 70
cm below surface to further explore this concentration. Excavation of the dark
circular depression continued until we reached the end of its defining dark brown
and reddish soil. Its total product can be viewed in the chart below (Table 1).
Artifacts from the dark circular depression contrast sharply with the mound centre
and KMK1. Yields from this unit peaked more than 50% at 30 to 70 cm below the
ground surface. An assortment of high quality ceramics, grinding stone fragment,
and lithics were recovered from H22's dark circular depression. A high

concentration of round lip rims and large body ceramic sherds were significant
finds. Unit H24 covers the southwestern continuation of the dark circular
depression. Because H24 was excavated to the end of the dark circular depression,
it has the greatest depth of the four units. Its ceramic frequencies are not
proportional to the more northern unit, H22-showing that this unit was on the
periphery of the concentrations associated with the dark circular depression.

Table 1: Units of Dark Circular Depression. Total Depth (cm below surface), and Total
Number of Artifacts

Unit Depth Artifact Count
H22 1.50 309
H24 1.80 170
122 1.40 101
124 90 288
Total I --- 868

Unit 122 is the far northeastern unit pertaining to the dark circular depression
and had the lowest frequency of artifacts. This unit was initially opened to cover
the northeastern span of the depression, but artifact recovery did not begin until 40
cm below surface and quickly culminated at 90 cm below surface, which may have
been due to sloping caused by its location on the mound edge. Unit 124 is located at
the southeastern portion of the dark circular depression. The frequency of
ceramics increased roughly 100% at approximately 20 cm below surface, but
quickly decreased by half at approximately 40 cm below surface. However, the
concentration of artifacts quickly shifted toward the unit's northeastern corner at
70 cm below surface where a 1 m x 1 m subsidiary unit was placed to focus on the
concentration. This general depth represents an increase in ceramic concentration
consistent throughout the dark circular depression, extending into the northern
wall of the H line. These findings caused us again to shift our attention on the dark
circular depression at the mound periphery. In that excavation, the artifact yields
peaked at 20 cm below surface and the yields were consistently high for the
remainder of its excavation to approximately 2 metres below surface. A potentially
high yield of artifacts remains to be excavated from within the dark circular

The large pot rims from the dark circular depression were identified by the
Saramaka crew as remnants of an ahgbang and/or bungu, large earthenware bowls
or pots used to boil and/or mix skillfully selected medicinal plants and other
powerful herbs for special ritual purposes. This is a common practice throughout
Surinamese Maroon culture. The mixtures of herbs are used to mitigate
troublesome issues concerning death, illness, and misfortune at the individual,
familial, and group level. It is believed that these issues are an act of an avenging lo
spirit in need of appeasement and acknowledgement (Price, 1983a, 1996). The
process entails the selection of choice, fresh vegetation mashed into a pulp, which
is then placed in a large ceramic pot where it is mixed with water and pemba.4
Washing occurs regularly in any given village (Figures 6a-b). The act of washing
(as it is called) with a selection of leaves precedes many ritual activities and
involves the mixing of specific types of vegetation with kaolin or pemba.
Moreover, when not addressing ancestral issues, washing can be interpreted as a
means of creating and maintaining a healthy deportment. This process of
purification and protection featuring pemba appears to draw strongly on African
symbolic values common to pemba most often associated with protection against
ancestral spirits and enemies, particularly in the Democratic Republic of Congo
and Angola. More widely, kaolin is used for the ritual washing of chiefs and kings
in central and east Africa during inauguration ceremonies and is also employed in
iron smelting rituals (Schmidt, 1997).
The mixture is scooped out with decorated calabashes and poured over the
individual's entire body several times. The curvilinear decorations in the
calabashes are a reflection of Saramaka artistic expression and can be found on
paddles, boats, houses, clothing, and even in the flat round food staple, cassava
bread. But these designs do not match the curvilinear design on ceramic artifacts
recovered from the site. The curvilinear designs found on contemporary Saramaka
objects have a broad and intertwined configuration. When making cassava bread,
women use both their index and middle finger to create a design that randomly
interconnects. The curvilinear motifs identified in the artifacts from the site are
made with a fine-tipped pointed object. Future ceramic finds with greater diversity
and detail in surface decoration may present an opportunity for more detailed
comparative analysis.
Ceramic pots, such as the ahgbang can be found located in a village's
goduosuu or prayer-ancestor shrine, along with a scrap ofpanghi cloth. Panghi is
a common article of clothing worn by all Maroon men, women, and children,
typically decorated in symmetrical patterns. In addition to the panghi cloth there is

a fist-sized ball ofpemba, and an empty rum bottle. Rum is a standard offering or
payment to an obeah man.5 As observed by Helman (1959) the nuances of the
goduosuu have been observed and described as:

The fetishes, which they cherish, decorated] with bits of cotton
and surrounded with offerings in the middle [middle] of the village
or in a hidden spot near the edge of the wood, lack all visual charm.
This for the reason that everything is literally charmed; the fetish is
only a symbol for recollection-the true spirit can never be
portrayed. Indeed, he will let himself be summoned and appeased
with gifts; indeed it is a good thing to smear oneself with white clay
in time of disease or trouble, for the evil spirit will fall back before
this change in appearance. It is better to respect, help and befriend
the village magician, the mediator between men and spirits. His
secret power is greater than that of the village's political head, who
is assisted by a council of elders and only speaks or is spoken to
through the mouth of a young 'speaker' during the long drawn out
meetings in the 'village hall' (Helman, 1959, 45).

Firul.'ri 6b: Ritual washing with
Figure 6a: Maroons engaging in a ritual washing r 6b: Ritual washing with
with choice leaves (Helman, 1959).pemba aki the first visit of
researchers C. White (on left) and
M. Heemskerk (on right) to the
Ndjuka village, Mooeitaki, on the
Tapanohony River in eastern
Suriname in 2003 (Photo by
Ndjuka villager).

All traditional or non-Christian villages host numerous visible goduosuu. In a
Christian village, these traditional shrines are placed more surreptitiously in the
centre living area of homes.
I can speculate that the concentration of ahgbangs from the dark circular
depression at KMK2's mound periphery marks the remains of an early Maroon
goduosuu. As I have observed, contemporary Maroons do not engage with the
actual goduosuu structures. They instead wash in an open area within the village. I
do, however, suggest that the concentration of ahgbangs may have existed as part
of an erected structure that may have been in a parallel location with an azonpow. 6
Its location at the mound periphery is at the mound's most accessible point from
the open area of KMK 1. This strategic placement alerts all visitors that they have
arrived at a community that recognizes and practices Maroon social and spiritual
culture. As a part of this structure, the ahgbangs may have served as washing
instruments for outsiders upon entering the protective enclave of KMK2's mound

Importantly, the KMK2 excavations on the mound periphery showed the
presence of ceramics that bear an uncanny likeness to indigenous (Carib) ceramics
that are today used in goduosuu shrines (Figure 7). This latter observation does not
mean that the same processes occurred over three hundred years ago, but the use of
indigenous artifacts certainly suggests a likely interaction with and utilization of
ideas and possibly material items derived from local populations.This observation

Figure 7: Goduosuu or prayer shrine (Price and Price, 1980).

is in keeping with similar conclusions drawn from archaeological studies in Brazil
(Allen, 1999, 2001), Cuba (La Rosa Corzo, 2003), and Florida (Weik, 2002). To
further understand what the remnants of ahgbangs may have represented to early
Saramaka, I examined what they mean to contemporary Saramaka.
I asked a series of questions about the acquisition ofahgbangs that appear in
traditional villages. The questions stemmed from an interest in learning how the
Saramaka differentiate domestic activity areas from an area of spiritual
significance: (1) What purpose does your goduosuu serve? (2) What do you do
with the ahgbangs found in the goduosuu? (3) How are they made? (4) What are
they made from? (5) What are they used for? (6) What meaning do they have?
Because of the specialization surrounding the uses of the ahgbang, the most
consistent responses were that it was a gift to the ancestors by the indigenous
people or that it is not for me to know. Price's (1983a) ethnographic accounts did
reveal that early Maroons developed relationships of a ritualistic nature with
indigenous peoples:
The intimacy of Indians and Africans in early Suriname society,
whether as fellow toilers on the plantations, hunters of runaway
slaves, or co-conspirators in rebellion, is confirmed again in the
matter-of-fact report that Kaasi had two wives in slavery-one
Indian, one African. And like Ayako, Kaasi always traveled with
his personal obiama, in his case the Indian Piyai (whose name
means "shaman" in the Indian languages of Suriname). Kaasi's
brother-in-law Piyai is said to have taught Kaasi dangara obia, the
magic that permits invisibility. The catching and marrying of an
Indian woman...is a theme that occurs in the traditions of other
clans as well, and must have happened a number of times. But the
crux of these fragments about Kaasi's adultery is that his act caused
an irrevocable division; his descendants, even today, continue to be
plagued by the avenging spirit (kunu) of Piyai, which serves as a
constant reminder of these domestic events that took place (Price,
1983a, 80).
This excerpt suggests that Maroons have a deep and enduring spiritual bond with
indigenous peoples of Suriname. To further the point, Price (1983a) stated that a
historical tale "alludes to other matters of historical interest as well-the
ambiguous role of Indians during the wars and the esoteric rituals of battlefield
death" (145). Price (1983a) interpreted that the presence of indigenous people
served more of a practical role in the lives of early Maroons: They aided travelling

clans in the early years in the forest: "They [indigenous women] made (often
reluctant) wives for several early Saramaka" and "they often served the whites as
scouts, guides, or bounty seekers, hunting down Maroons for cash" (145).
Whether or not the interaction of indigenous peoples and early Maroons is directly
related to the appearance of ceramic vessels in a goduousu is not clear, but the
interaction requires examination of possible indigenous influences. Early
Maroons surely would have found the "magic that permits invisibility" to be useful
to their cause in maintaining the secrecy of their villages, besides offering a
strategic advantage during bouts of guerilla warfare with Dutch militia. Though
Maroons reside openly and freely along river-based villages today, they may still
have a need to appease these early historical figures that played such an important
role in the maintenance of their identity while developing Maroon clans and
isolated settlements. Because collecting information about the meaning of these
shrines proved to be problematic, this forced the emphasis of this research more
toward the material resource used.
To observe the construction of ahgbangs and in order to ascertain an
authentic Saramaka technology through informal interviews and participant
observation, I made an attempt to interview Saramaka potters regarding the
technique of manufacturing and their material source. But after questioning elders
from non-Christian villages (Tchjali Kondre, Goejaba, and Pikein Slee), about the
retention of this knowledge, I learned that only one living woman still maintained
the knowledge of pottery construction. At the onset of the August 2004 project
season, this woman's husband had just died, thus requiring a mandatory six-month
mourning period and barring any further engagement regarding the technology of
pottery. To address the resource acquisition question from another perspective I
explored the circumference of the mound in search for anomalies in the terrain,
such as rock outcrops or dried creek beds that could have been a source for clay
extraction. I believed that this approach would offer insight to geographical and/or
geological differences that might differ from the dense tropical forest. The
circumference of the mound was surveyed using the demarcating ditch as the
starting point for 5 metre transects that extended 15 to 20 metres beyond the ditch
(the dense vegetation created some difficulty in maintaining a consistent twenty
metres). Several potential anomalies on the ground surface were identified, and
tested, but did not produce artifacts to warrant further exploration. Attention was
instead focused on KMK2 mound centre.

KMK2 mound centre
The entire periphery of the mound is covered with thick brush and
old-growth vegetation (Figure 8a-b). The centre of mound, however, did not have
old-growth vegetation. This distinction leads to the interior of the mound being
identified as the village centre (The term 'mound centre' refers to the physical
description while the term 'village centre' is the interpretative description,
hereafter the terms village centre and mound centre are used interchangeably). The
village centre lacks the dense, high forest canopy indicative of the surrounding
tropical forest, though it did have scattered vegetation up to approximately 20
metres from the ground surface. The absence of the canopy allowed for a full view
of the sky. Figure 8c shows a cross-section of the mound and the relationship of the
village centre vegetation to its high peripheral canopy.
To test this part of the site, a (large) 1 m x 2 m shovel test pit was placed in
the mound's village centre, made accessible by clearing. The shovel test pit
produced a considerable amount of earthenware recovered from the first 20 cm
below surface and further justified investigation of the village centre. The soil
colour tested as Munsell 1 OYR 4/6 (dark yellowish brown), dry and compact. A 50
m x 50 m area of the village centre was then cleared using a slash-and-bum
technique. An excavation grid consisting of 2 m x 2 m units covered 16 square
metres within this area. We excavated two units located at the north and south
extremity of the grid to sample the village centre. To date, the entire village centre
has not been cleared or further excavated. Future research will enlighten us as to
the characteristics and cultural attributes of the village centre.
The artifacts recovered from the KMK2 mound centre produced a different
type of assemblage from that which had been recovered in the previous year's
excavations of KMK1. We found a greater variety in ceramic forms and lithic
artifacts from the KMK2 village centre of Kumako. Distinctive of these
assemblages are rim sherds, bases, handles, large body sherds, and ground stone
and projectile point fragments. The distinctive artifacts in the mound centre also
contrast with the mound periphery, which has greater numbers of large rim and
body sherds.
The excavation of the mound periphery extended further below surface than
in the village centre. Although only two units have been sampled from the village
centre, the concentration of artifacts below the surface was greater and contained
more ceramic and lithic variance than KMK1, suggesting a greater diversity of
activities, such as a limited craft specialization in the lithic assemblage. While the
village centre units were not excavated to sterile soil due to time constraints, it

appears from our field observations that the concentrations were within the first 20
cm below surface. The two units plus the village centre shovel test pit produced the
highest yield at 10-20 cm below surface, with a combined count of 399 sherds, or
99.4% of the ceramics. This strong bias-where there is more diversity in other
parts of the site-may well point to some spatially defined activity area where
ceramic use was privileged, be it for social or other reasons.

Figure 8a-c. The centre of Kumako 2 mound or village centre with a visible sky and the
surrounding old-growth forest along the periphery in the back. Figure 8a and 8b (photos
by M. Heemskerk 2004).

rig aa rig aD

High forest
canopy of


Village Center -


Mound Cross-section


Fig 8c

Intra-site Differentiation at Kumako
Research questions were addressed based on key observations about
differences in intra-site distributions of artifacts found at various sections of
Kumako. As shown in COLOUR PLATE 5 p.vi), KMK1 produced a higher yield
of non-ceramic artifacts, such as lithics of ground stone, projectile point
fragments, and bifacial flakes. The metal objects included nails and musket balls,
and the organic remains consisted of cowry shells and charcoal. Kumako2,
however, presented a larger yield of ceramics in the form of assorted rims and
body sherds. Taking a closer look at COLOUR PLATE 5 ( p.vi) we see that the
most varied distribution of ceramic and non-ceramic artifacts is at KMKI.
However, the highest frequencies of ceramics and lithics are at KMK2.
This dichotomy speaks to intra-site distributions suggestive of a settlement
arrangement at Kumako. Moreover, intra-site distribution of artifacts reflects
structured choices for areas of activity. COLOUR PLATE 6 (p.vi) shows the
distribution of different parts of ceramic vessels across KMK1 and KMK2.
Kumako2 has a higher frequency of morphologically distinct ceramics:
bases, handles, and curved body sherds. Most of these ceramics come from the
highly specialized dark circular depression at the mound periphery, and they are
likely linked to ritual activity-a very likely location for communal rituals at a
central locale or plaza.
A higher frequency of unidentified (UID) sherds and fewer types of body
sherds appeared at KMK1. This juxtaposition of artifact assemblages at KMKI
(greater preponderance of non-pottery items and unidentifiable sherds) and at
KMK2 (increased yield of identifiable rim and body sherds) leads us to speculate
that the Kumako settlement in its entirety may have had binary functions or
distinctions, perhaps linked to the different transitory social groups who would
later develop into the Matawai tribe.
A likely proposal is that KMK2's village centre may have been a locale for
conducting domestic activities in a non-exposed environment. The KMKI
perimetre may have presented a buffer zone or defensive barrier, or perhaps even a
cultivation area for ground provisions. The tall, dense tropical vegetation that
outlines the mound periphery creates an ideal protective enclave for the village
centre by giving maximum protective advantage to its inhabitants. The sharp
slopes of the mound periphery also create a visual advantage for residents looking
out from the interior, allowing inhabitants to assess activity beyond the mound
without being noticed. In addition, this earthwork also prevents intruders from
observing the nucleus of the Saramakaans' first permanent settlement.

Kumako was an early settlement for the Saramaka. The intent of the initial
research questions regarding kin-group divisions and clusterings, and material
culture resources was to explore the nuances of the settlement. The material
remains suggest that KMKI was used for domestic activity while KMK2 mound
periphery was used for ritual activity, and the mound centre was reserved as a
protective enclave. More archaeological investigation is needed to understand
how Kumako and Tuido may have functioned in unison, while enabling Maroons
to culturally flourish in isolation.

1. Elevation is based on the average elevation within the general settlement and is
derived from five elevation points on site periphery.
2. These months represent Suriname's dry season and therefore the best time of year
to travel and work in the forest proper.
3. This Matawai elder died shortly after alerting Dr. Agorsah and other members of
the MHRP team of the location of Tuido. He was purported to be the last living
individual to be born at Tuido.
4. Pemba is a widespread Bantu word that specifically applies to kaolin used in ritual
5. The term obeah refers to Maroon spiritual practice. It entails ancestor worship and
appeasement through the form of food offerings and communal acknowledgement of
group issues and problems. Obeah is an African-derived spirituality practiced among
many peoples of African descent throughout the New World.
6. An azonpow is a wooden rod placed horizontally at the entrance of a traditional
Saramaka village. It demarcates a village that practices traditional rituals.

Allen, S. (1999). Africanism, mosaics, and creativity: The historical archaeology of
Palamares. In P. P. A. Funari (Ed.), Cultura Material e Arqueologia Historica (pp.
141-178). Campinas: IFCH/UNICAMP.
Allen, S. (2001). 'Zumbi Nunca Vai Morrer': History, the practice of archaeology, and
race politics in Brazil. Unpublished PhD dissertation, Department of Anthropology,
Brown University, RI.
Bos, G. (1998). Some recoveries in Guiana Indian ethnohistory. Amsterdam: Vrije
Dillon, B. D. (Ed.). (1993). Practical archaeology: Field and laboratory techniques and
archaeological logistics. Los Angeles, CA: Institute of Archaeology.

Fermin, P. (1781 [1778]). Historical and political view of the present and ancient state of
the colony ofSuriname in South America; with the settlements ofDemerary and
Issequibo; together with an account of its produce for 25 years past. London: W. Nicoll.
Goslinga, C. (1971). The Dutch in the Caribbean and on the Wild Coast 1580-1680.
Gainesville, FL: University of Florida Press.
Goslinga, C. (1979). A short history of the Netherlands Antilles and Surinam. London:
Martinus Nijhoff.
Goslinga, C. (1985). The Dutch in the Caribbean and in the Guianas 1680-1791. The
Netherlands: Van Gorcum.
Green, E. (1974). The Matawai Maroons: An acculturating Afro-American society.
Anthropology Studies No. 28. PhD dissertation, The Catholic University of America.
Helman, A. (1959 [1957]). Suriname: A new nation in South America. James Healy,
trans. Paramaribo: Radhakishun.
Hoogbergen, W. (1990). The Boni Maroon wars in Suriname. Leiden: E. J. Brill.
Jackson, M. (1965). Salt, sugar and slaves: The Dutch in the Caribbean. James Ford Bell
Lecture, No. 2. Minneapolis MN: Associates of the James Ford Bell Collection.
Khudabux, M. (1999). Effects of life conditions on the health of a Negro slave
community in Suriname. In J. B. Haviser (Ed.), African Sites: Archaeology in the
Caribbean (pp. 291-312). Princeton, NJ: Marcus Weiner Publishers.

Klooster, W. (1997). The Dutch in the Americas, 1600-1800: A narrative history with a
catalogue of rare prints, maps and illustrated books from the John Carter Brown
Library. Providence, RI: The John Carter Brown Library.
La Rosa Corzo, G. (2003 [ 1988]). Runaway slave settlements in Cuba: Resistance and
repression. Mary Todd, trans. Chapel Hill, NC: The University of North Carolina Press.
McMillon, B. (1991). The archaeology handbook: A field manual and resource guide.
New York: John Wiley and Sons, Inc.
Price, R. (1975 [1974]). Saramaka social structure: An analysis of a Maroon society in
Surinam. Caribbean Monograph Series, No.12, Institute of Caribbean Studies. Puerto
Rico: University of Puerto Rico.
Price, R. (1983a). First-Time: The historical vision of an Afro-American people.
Baltimore, MD: The John Hopkins University Press.
Price, R. (1983b). To slay the hydra: Dutch colonial perspectives on the Saramaka Wars.
Ann Arbor, MI: Karoma Publishers.
Price, R. (1990). Alabi's world. Baltimore, MD: The John Hopkins University Press.
Price, R. (Ed.). (1996). Maroon societies: Rebel slave communities in the Americas. 3rd
edition. Baltimore, MD: The John Hopkins University Press.
Price, R. and Price, S. (1980). Afro-American arts of the Suriname rainforest. Berkeley,
CA: University of California Press.

Price, R. and Price, S. (1988 [1790]). John Gabriel Stedman: Narrative ofa five years
expedition against the revolted Negroes of Surinam. Manuscript transcribed for the first
time by Richard and Sally Price. Baltimore, MD: The John Hopkins University Press.
Price, R. and Price, S. (1991). Two evenings in Saramaka. Chicago, IL: The University
of Chicago Press.
Price, R. and Price, S. (1992). Stedman 's Surinam: Life in an eighteenth-century slave
society. Baltimore, MD: The John Hopkins University Press.
Price, R. and Price, S. (2003). The roots of roots: or how Afro-American anthropology
got its start. Chicago, IL: Prickly Paradigm Press.
Schmidt, P. (1997). Iron technology in East Africa: Symbolism, science, and
archaeology. Bloomington, IN: Indiana University Press.
Stedman, J. G. (1796). A narrative of five years' expedition against the revolted
negroes ofSurinam, in Guiana, on the wild coast of South America, from the year 1772
to 1779. London: J. Johnson and J. Edwards.
Weik, T. (2002). A historical archaeology of Black Seminole Maroons in Florida:
Ethnogenesis and culture contact at Pilaklikaha. Unpublished PhD dissertation,
Gainesville, FL: University of Florida Press.

Linstead Market before Linstead?
Eighteenth-century Yabbas and the Internal
Market System of Jamaica


The role of Linstead market in the internal economy of post-emancipation
Jamaica has been the subject of numerous popular and academic discussions.
Indeed there is ample documentary evidence that points to its existence at least as
early as the 1840s. Any arguments regarding its existence in the eighteenth century
at this point is speculation based on models of economic necessity, our
understanding of the organization of enslaved labour revolts, and demographic
probabilities. At the time the area in which Linstead Market would be based was
called Sixteen-Mile-Walk and was home to numerous plantations. The difficulty
in trying to ascertain the organization of such markets in the eighteenth century is
that they were inevitably poorly documented with only fragmentary accounts. In
this paper I attempt to antedate the Linstead Market using multiple sources of
evidence including, published accounts, unpublished ethnography and
archaeology. Specifically, analysis of production and distribution ofyabba, a local
Jamaican ceramic, suggests an eighteenth-century establishment of Linstead
Market, thus tying this space to one of the significant slave rebellions of
eighteenth-century Jamaica.
Markets and Unrest
By the mid-eighteenth century, the internal market system figured centrally
in Jamaican economic and social lives. The social importance of the combination
of street markets, itinerant sellers, and small-scale trade is one of the reasons why
such institutions have figured so prominently in the historiography of the African
Diaspora. As many have noted the internal economy was a locus of independent
acquisition, marketing, and production among the enslaved (Hall, N., 1977; Hall,
D., 1980, 1985, 1994; Bush, 1981; Simmonds, 1987, 2004; Tomich, 1995; Boa,
1993; see the volume edited by Gaspar and Hine, 1996; Hall 1999). This economy
also presaged a Caribbean peasantry rooted in the houseyard and market (Hall,
1959; Mintz, [1974] 1992; Craton, 1982; Trouillot, 1988). In the eighteenth
century, few other institutions were as explicitly impacted by the rural and urban

freed and enslaved. Everybody in Jamaica was dependent on the internal
economy, some to a greater extent than others.
The independent production by enslaved labourers on provision grounds
and the exchange of those goods were activities that found constructed spaces on
the margins of the planters' figurative and material control (Pulsipher, 1986, 1990,
1991, 1994; McKee, 1999; Pulsipher and Goodwin, 1999). Consequently, the
informal markets as a meeting place of goods and ideas of the enslaved can be
viewed as a locus of interaction where the enslaved could transgress the social and
geographic boundaries imposed by the plantation. It is in these markets that we see
a struggle over the valuation of local goods, to prevent engrossing the price of
staples, but also enumerating the perpetrators of market disorder as "Indian,
Mulatto, or Negroe."' Planters attempted to circumscribe market participation
through a series of legislative mechanisms (see Mintz and Hall, 1991; Simmonds,
1987, 2004; Hauser, 2008). Planters were worried about two things, theft and
association. The legal code monitored these two threats through a system of tickets
and surveillance (Hauser, 2008, 56).
As Barry Higman has noted in his analysis of documents associated with the
artifacts recovered from Montpelier estate, that community emerged out of a
shared sense of locality, kinship, language, values and reciprocity. More
importantly, the people living at Montpelier were not confined geographically to
the boundaries of the estate, rather their social landscape extended beyond the
estate grounds through direct and indirect means (Higman, 1998, 297-305). By
way of example, he cites the Baptist War, the 1831/32 slave rebellion which
consumed the western parishes of Jamaica. He highlights that for the individuals
responsible for its conception and execution, the markets provided one locus for
planning in which information could be passed and the uprising organized
(Higman, 1998, 262-263).

It is possible that riots, uprisings, and rebellions in the eighteenth century
were also organized through the social circuitry of the internal market system. This
question first became apparent to me when I read an excerpt of a letter written in
1765 by Simon Taylor to a friend in London:
We about a fortnight ago had an Alarm of Rebellion in St. Mary's
when Matt. Byndloss and my Overseer were both murdered by a
parcel of new Negroes belonging to the Overseer of Whitehall and
Ballards Valley...Another of them was taken up in this Town who
has impeached all the Coromantees on Albion Trinity and the
Frontier and their design was to have broke out a Month after

Christmas and to have attack the fort at Port Maria to gett arms and
powder and from thence to go to Sixteen mile walk where there
were many of their countrymen and that the Negroes of Scots Hall
were to have joined them. That it broke out was occasioned by the
New Negroes declaring that they would wait no [longer]. There is
report that there was to have been some disturbances at
Westmoreland (Simon Taylor, December 9, 1765, in Wood &
Martin, 2002, 30).

The places Taylor's letter identify were affected by the action as well as Port Maria
and Sixteen-Mile-Walk. What is important about Sixteen-Mile-Walk? Why
would the 'Coromantees' decide to meet there?
The area around Sixteen-Mile-Walk later came to be known as Linstead
Market. While we know Linstead Market was in operation at least as early as
emancipation (Robertson, 2005, 200), the above quote suggests there might have
been a market there prior to emancipation, although on a less formal basis. It is also
interesting to note that one plausible candidate for a source of pottery manufacture
continues to be the area around Sixteen-Mile-Walk. This would potentially enable
us to locate in place and time a market feared to have existed, but that was
ultimately outside of Simon Taylor's knowledge. By the mid nineteenth century,
however, this market becomes "established" when a visiting Governor described it
as a newly established township (Robertson, 2005, 200).
Archaeology, with its focus on material culture and the distribution of
material goods both within and between excavated communities of past peoples
can provide a tool through which to establish potential sites of social and economic
interaction. Indeed, rather than remaining at the level of descriptive analysis,
archaeologists attempt to apply explanatory devices to determine the processes
through which artifacts find themselves in the kitchens of the elites and the
houseyards of labourers. In the case provided here, by looking at one form of
material culture, the Jamaican yabba, and determining their sites of use along with
potential loci of production, we can begin perhaps to recognize circuits of trade
through which commodities, both local and imported flowed. Given the
importance of independent production among enslaved labourers in the provision
grounds and the houseyard, and their reliance of the system of street markets and
higglers, the internal economy of Jamaica has become an important explanatory
device to understand the material life of the enslaved.