The Florida anthropologist

Material Information

The Florida anthropologist
Abbreviated Title:
Fla. anthropol.
Florida Anthropological Society
Place of Publication:
Florida Anthropological Society.
Publication Date:
Quarterly[<Mar. 1975- >]
Two no. a year[ FORMER 1948-]
Physical Description:
v. : ill. ; 24 cm.


Subjects / Keywords:
Indians of North America -- Antiquities -- Periodicals -- Florida ( lcsh )
Antiquities -- Periodicals -- Florida ( lcsh )
serial ( sobekcm )
periodical ( marcgt )


Contains papers of the Annual Conference on Historic Site Archeology.
Dates or Sequential Designation:
v. 1- May 1948-

Record Information

Source Institution:
University of Florida
Holding Location:
Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
01569447 ( OCLC )
56028409 ( LCCN )
0015-3893 ( ISSN )

Full Text


2000 Florida Anthropological Society Inc.

The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
these publications.

all rights to the
and shall be
and images of

The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.

The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.

Contact the Florida Anthropological Society for additional
information and permissions.

E FL ID k9 IID*4

1 U4 vs, u


JAN _1 9


ce, &goo

Ir C H AXXjr,





THE FLORIDA ANTHROPOLOGIST is published quarterly in March, June,
September, and December by the Florida Anthropological Society, Inc.,
Geoarcheological Research Center, Department of Geology, University
of Miami, Coral Gables, Fl 33124. Subscription is by membership
in the Society for individuals and institutions interested in the
aims of the Society. Annual dues are $8.00; student members $5.00.
Requests for membership and general inquiries should be addressed to
the Secretary; dues, changes of address to the Treasurer; manuscripts
for publication to the Editor; orders for back issues to the Assistant
Editor; and newsletter items to the President. Address changes should
be made at least 30 days prior to the mailing of the next issue.
Second class postage paid at Miami, Florida 33124.


President: Thomas Watson
3705 Dilwood Drive
Panama City, FL 32407

1st Vice President: Irving Eyster
Route 1, Box 96
Islamorada, FL 33036

2nd Vice President: Stephen C. Atkins
P. 0. Box 660
Micanopy, FL 33627

Secretary: Marion M. Almy
5321 Avenida del Mare
Sarasota, FL 33581

Membership Secretary: Jeane Eyster
Route 1, Box 96
Islamorada, FL 33036

Treasurer and Resident
Agent: Larry Hochen
P. 0. Box 330754 Coconut Grove
Miami, FL 33133


Three years: Karen Malesky
5312 Bay State Road
Palmetto, FL 33561

Two years: Norcott Henriquez
1510 Dewey Street
Hollywood, FL 33020

One year: Adelaide Bullen
Florida State Museum
Gainesville, FL 32611


Editor: Robert S. Carr
Geoarcheological Research Center
Department of Geology
University of Miami
Coral Gables, FL 33124

Newsletter Editor: Wilma Williams
2511 McKinley Street
Hollywood, FL 33020

Editorial Board:
Kathleen A. Deagan
Department of Anthropology
Florida State University

John W. Griffin
St. Augustine, Florida

CUSPS 200880)

Assistant Editor: Irving Eyster
Route 1, Box 96
Islamorada, FL 33036

Guest Editor: John Winter
Molloy College
Rockville Centre, N.Y. 11570

George M. Luer
Sarasota, Florida

James J. Miller
Tallahassee, Florida

George Percy
Div. of Archives, History and
Records Management, Tallahassee

COVER: From A General Topography of North America and the West Indies,
1768, by Thomas Jefferys, (1763). Courtesy of Historical
Association of Southern Florida.







Editor's Page . . . .

Preface . . . . .

Second Bahamas Conference Participants .

A Brief History of Bahamian Archaeology
by.Julian Granberry . . .
The Concept of Series in Bahamian Archaeology
by Irving Rouse . . .

The Paleoecology of the Caribbean Area
by Victor A. Carbone . . .

An Overview of the 1976 to 1978 Archeological
Investigations in the Caicos Islands
by Shaun D. Sullivan . .

The Trade Process and the Implications of
Trade in the Bahamas
by Richard E. Daggett . .

Aboriginal Trinidad in the Sixteenth Century
by Stephen D. Glazier . . .


. 79

. 80

S. 82

. 83

S. 94

. 99

S. 120

S. 143




This issue features papers presented at the Second Bahamas
Conference on Caribbean prehistory that was held in Rockville Centre,
New York in October, 1978. John Winter, one of the principal organ-
izers of the conference co-edited this issue, and I am very grateful
for his help and his decision to publish these proceedings in the
Florida Anthropologist. I also want to thank John Beriault for his
comments and George Luer for drafting the map for the Julian Cranberry

The Florida Anthropologist will continue to publish on occasion
special thematic issues. Special issues in the coming months will
feature the Archaic Period in Florida and another issue on the
Seminoles. Anyone who has contributions for these issues is en-
couraged to submit them as soon as possible.

Florida Archaeology, a book written by Drs. Jerald T. Milanich
and Charles H. Faribanks, has just been published by Academic Press
and is now available. The book provides an up-to-date synthesis of
Florida archeology, and is written specifically for students, amateur
archeologists, and interested laypersons. This book is hard bound; has
330 pp., and may be ordered from Academic Press, Order Department,
111 Fifth Avenue, N.Y., N.Y. 10003 for $19.50.

Readers will be interested in knowing that a new organization,
the Archeological and Historical Conservancy, has just been incorporated
Modeled after the Nature Conservancy, this organization will attempt to
preserve significant archeological sites in Florida through tax exempt
donations and purchases. Regular membership is available for $10.00
and family membership is $25.00. Members will receive a quarterly pub-
lication, Current Projects, that will provide information on ongoing
archeological research, archeological surveys, excavations, and
preservation grants in the State of Florida by governmental agencies,
universities, contractors, archeological societies, and both professions
and non-professional archeologists.

The Conservancy's officers are: Irving Eyster, president; Robert
Carr, vice president and editor; and John Beriault, acting treasurer.
Memberships and additional information can be secured by writing to
P. O. Box 450283, Miami, Florida 33145.

The Florida Anthropologist still needs papers, particularly from
non-professionals. I would like to discourage contributions that
exceed 44 typewritten pages (including plates) unless the paper pre-
sents some new and significant information.



From June 11-15, 1966, the First Bahamas Conference on Arche-
ology was held at the New World Museum on San Salvador, Bahamas,
under the sponsorship of the Center for Latin American Studies, of the
University of Florida. The purpose of the Conference was to study
the problems in Bahamian prehistory and the relationship of the
Bahamas to the Caribbean. The presentations were based upon arche-
ological research and ceramic typology of the Greater and Lesser
Antilles, and the Bahamas region. An area of discussion also cen-
tered upon the role of the small museum in Caribbean research.
Some of the original participants at that first conference were
Ripley and Adelaide Bullen, William Haag, Charles Hoffman, Jr.,
Stephen Gluckman, Fred Olsen, Irving Rouse, Ronald Vanderwal,
Elizabeth Wing, and Ruth Wolper Malvin.

Since that conference, three archeological investigations have
been conducted within the Bahamas region: Shaun Sullivan (Univer-
sity of Illinois) on Eleuthera and the Turks and Caicos; Marjorie
Pratt (College Center of the Finger Lakes, New York) on San Salvador;
and John Winter (Molloy College) on Crooked Island. As a result of
this expanded research within the area, I spoke with Drs. Julian
Granberry and Irving Rouse about the feasibility of holding a second
Bahamas conference. I proposed that the Second Bahamas Conference
on Archeology be held on October 13-14, 1978, at Molloy College,
Rockville Centre, New York. Letters were sent out to all the parti-
cipants of the first conference, and to a limited number of those
individuals who were known to have an interest in the Bahamas and
the Caribbean region.

Fourteen papers and a workshop on ceramic classification were
scheduled for the two day conference. The majority of the presen-
tations dealt with theoretical models of Caribbean migration and
ecology, archeological research, and ceramic typology for the
Greater Antilles and the Bahamas region. Other presentations covered
the archeology of Venezuela and the ethno-history of Trinidad. Al-
though fourteen papers were presented, only six are being published
in this journal. This is because some individuals either never sub-
mitted a final copy or believed that their paper needed more research
before it could be published.

One of the major outcomes of this conference, as with many small
specialized conferences, was that individuals were able to meet with
people whom they may never have met at larger conferences. It also
allowed individuals from the first conference to become re-acquainted
after 12 years. Moreover, it gave the individuals the opportunity to
contemplate and discuss the new information that had been gathered
relative to Bahamian prehistory. It was unfortunate that no decision
was reached as to which ceramic typology is best to use for the
Bahamas, (see Rouse's article in this volume) but the conference
did point out the need to obtain additional radiocarbon dates in
the Bahamas to provide an accurate picture of the chronology of pre-
historic migrations in the Bahamas relative to other parts of the

Caribbean. The members present tentatively decided that a third
conference should be held in 1982, at the New World Museum on San

I would like to thank S. Janet Fitzgerald, President, and S.
Katherine Gee, Chairperson of the Sociology, Anthropology and Social
Work Department for giving me permission to hold the conference at
Molloy College. I would also like to thank Robert Carr, editor of
the Florida Anthropologist, for providing me with a vehicle through
which to publish a portion of the Proceedings. I would also like to
thank those authors, who submitted manuscripts, with the anticipa-
tion of an "early" publication for the Proceedings. I would also
like to thank the participants for making the Second Bahamas Confer-
ence on Archeology the success it was and to Alfredo Figueredo who
kept encouraging me to make the conference a reality.

John Winter
Molloy College
Rockville Centre, New York


Juan Jose Ortiz Aguilu
Dept. of Archaeological Research
Museo Del Indio Antilland
Robert Stolberg-Acosta Foundation
265 South 21 Street
Philadelphia, Pennsylvania 19103

Richard E. Daggett
Dept. of Anthropology
University of Massachusetts
Amherst, Massachusetts 01003

Alfredo E. Figueredo
Virgin Islands Arch. Society
P.O. Box 368
St. Croix, U.S.
Virgin Islands 00840

Stephen D. Glazier
Dept. of Anthropology, U-158
University of Connecticut
Storrs, Connecticut 06268

Julian Granberry
Dept. of Anthropology
St. John Fisher College
Rochester, New York 14624

Bill Keegan
427 Parker Street
Manchester, Connecticut 06040

Elizabeth Kijourg
170 Old Quarry Road
Guilford, Connecticut 06437

Barbara S. Macnider
370 Forest Trail
Oakbrook, Illinois 60521

Ruth Durlacher Malvin
7811 North Shore Road
Norfolk, Virginia 23505
New World Museum
San Salvador Island

Marisol Melendez
229 East 10th Street, Apt. 3
New York, New York 10003

Fred Olsen
Box 303
Guilford, Connecticut 06437

Marjorie K. Pratt
1220 Euclid Avenue
Syracuse, New York 13210

Andres M. Principe
1901 Spruce Street, Apt. 1
Philadelphia, Pennsylvania 19103
Box 106
Yauco, Puerto Rico 00768

Lydia M. Pulsipher
422 Cortlandt Avenue
Mamaroneck, New York 10543

Irving Rouse
Dept. of Anthropology
Yale University
New Haven, Connecticut 06520

Riva Berleant-Schiller
25 Highfield Road
Glen Cove, New York 11542

Shaun Sullivan
6500 Chesterfield Avenue
MeLean, Virginia 22101

Julia Tuares
230 Prospect Street
New Haven, Connecticut 06511

Gary S. Vescelius
Virgin Islands Office of
Archaeological Services
Box 7818
Charlotte Amalie, Virgin Islands

John H. Winter
Dept. of Sociology, Anthropology
and Social Work
Molloy College
Rockville Centre, New York 11570


Julian Granberry

Archeological investigations in the Bahama, Turks, and Caicos
Islands, which will, on cultural and geographic grounds, be con-
sidered here a single archipelago, date back to the brief account
by W.K. Brooks in 1888 of the discovery of three Lucayan crania.
It has resulted in eleven fairly extensive collections of archeo-
logical material in the United States as well as smaller ones both
in this country and abroad. It was an examination of these col-
lections that led to the first attempt at a tentative synthesis of
cultural data and an accompanying equally tentative diachronic re-
construction of Lucayan culture (Granberry, 1952, 1955, 1956), and
it has presumably been, at least in part, this synthesis, in turn,
which has helped give rise to the increasingly more frequent archeo-
logical work in the archipelago to the present.

The earliest, and still by far the most important, work done in
the Bahamas were the surveys made by W.K. Brooks, Theodoor de Booy,
Froelich G. Rainey, John M. Goggin, and Herbert W. Krieger. These
surveys yielded copious surface collections from almost all parts of
the islands, but the only excavation undertaken was Rainey's work at
Gordon Hill, Crooked Island, and presumed excavations on various
islands by Krieger. Unfortunately we do not have Mr. Krieger's
field notes, and much of his actual data seems to have vanished, so
it is as yet not possible to determine the exact extent of his exca-

The first published account on Bahamian archeology is the above-
mentioned paper by Brooks in the Memoirs of the National Academy of
Sciences (vol. 4, 1888). It consists of a description of three
crania found in caves in various parts of the archipelago, exact pro-
venience unknown. Two of the crania were adult males, one an adult

The female skull was given to the Morton Collection of Crania
Americana in Philadelphia by the owner, Dr. J.C. Albury of Nassau.
It lacks the lower mandible, but is otherwise perfect. The first
male skull comes from a cave in one of the Out Islands; we do not
have even that amount of information on the third skull. Both of
the male skulls are complete with lower mandibles. The one refer-
red to by Brooks as No. 3 was in poor condition when he examined it.
At that time both male skulls were in the Public Library in Nassau.
Shattuck (1905: 421) mentions a single skull from the library, which
he illustrates with a photograph. Comparison of Shattuck's photo-
graph with the plates and data in Brooks' account indicate it to be
Brooks' Skull No. 2.

Brooks also examined some fragmentary skeletal materials from
the collection of Edith, Lady Blake, wife of the governor of the

Bahama Litte Abaco


SThe ~ Berry Is lan
o* -^

300 km
p p *




Cat Island
0 San Salvador
inception Is.o Rum Cay
Exumaw Cay
Long Is and

Crooked Is. Pan
Fortune Is. ana
Ragged ck linsc==.Mayaguana
Is .o Is. Norl


th Caicos

Providencialesc: <0 Middle Caicos
o 'East Caicos
Grand Turk




Major islands of the Bahama Islands, Caicos Islands, and Turks Islands.



*Cay Sal



Figure 1 0


Bahamas. Lady Blake had in her collection portions of two skele-
tons from a cave on New Providence, including the roof of a cranium
with the frontals and parietals nearly complete, together with part
of an occiput and broken maxillae and malars, probably from the same
skeleton; the frontal and fragments of the parietals, occipitals, and
mandible of a second cranium; three femurs, three radii, three fibu-
lae; an innominate bone; a sacrum; fragments of two or more humerii;
several vertebrae; and a number of fragments of various bones (Brooks
1888: 216).

From June to December, 1912, Theodoor de Booy and his assistant,
Mr. C.V. Spicer, conducted an archeological survey of the Bahamas,
Turks, and Caicos, under the auspices of George G. Heye, founder of
the Museum of the American Indian in New York City. This represented
the first survey made of the archipelago and remains to this day by
far and away the most thorough and important survey made of the archi-
pelago. Results were published in two reports, appearing in the
American Anthropologist (De Booy, 1912, 1913). Mr. De Booy explored
and collected material from some thirty-eight sites in all, but he
apparently did no actual excavation. Among the more important sites
he investigated were Juba Point, West Harbour Bluff, Indian Hill
(Providenciales); Sandy Point, Pumpkin Bluff, Whitby, Bellevue Mounds,
Kew (North Caicos); Ferguson's Point, Conch Bar, Lorimers, Bambara
(Grand or Middle Caicos); and Jacksonville (East Caicos). Various
sites were visited in the Turks group. In the Bahamas proper he in-
vestigated Mayaguana, Eastern Plana Cay, the Biminis, Cat Island, Long
Island, Eleuthera, Little Abaco, San Salvador, Great Ragged Island,
Andros, Rum Cay, Acklins Island, Crooked Island, New Providence, and
Great Inagua. As can be judged, his survey was extensive. It is un-
fortunate that the only reports published were rather brief. However,
he gathered much material from these sites, all of which is at the
Museum of the American Indian.

During February and March, 1934, Froelich G. Rainey, then a
graduate student at Yale University, conducted an archeological survey
of the Bahamas on behalf of the Yale Peabody Museum. This expedition,
part of Yale's Caribbean program, was made possible by Mr. Allison V.
Armour of New York, who invited Rainey to accompany him on his re-
search yacht Utowana for a general survey of archeological sites in
the Bahamas and Haiti (Rainey 1941: 3). A brief account of the sur-
vey is contained in vol. XVIII, part 1, of the Scientific Survey of
Porto Rico and the Virgin Islands (Rainey 1940). Additional infor-
mation on Dr. Rainey's work is available in his field notes (Rainey
1934), which are on file in manuscript form at the Yale Peabody Muse-
um, New Haven. Dr. Rainey undertook investigations on eleven islands,
including Grand Bahama, Great Abaco, New Providence, Eleuthera, Cat
Island, Conception Island, Rum Cay, San Salvador, Long Island,
Crooked Island, and Great Inagua. Other islands, including Mira Por
Vos and the Fish Cays, were visited briefly during the course of his
research. In all, fifteen productive sites were found. Much of
Rainey's material, including most of the stone celts, was purchased
on various islands. The expedition was primarily a reconnaise one,
and only one site was thoroughly excavated, the Gordon Hill site on


Crooked Island. The majority of the sites were cave burials; none
were open village sites. In cases where culture deposits were pre-
sent, material was extremely rare and quite fragmentary a total of
only 373 ceramic specimens are included in his collection, all con-
sisting of small sherds.

From October, 1936, to February, 1937, Herbert W. Krieger of
the United States National Museum in Washington conducted an archeo-
logical reconnaissance of the Bahamas. A preliminary report was pub-
lished in the Explorations and Field Work of the Smithsonian Insti-
tution in 1936 (Krieger 1937), and brief mention of this expedition
is made in the Annual Report of the Smithsonian Institution for 1937
(Smithsonian Institution 1938). Mr. Krieger visited eight of the
islands, including New Providence, Eleuthera, Long Island, Cat
Island, San Salvador, Great Inagua, Andros, and the Berry Islands.
The only specific sites mentioned in the preliminary report are
Hamilton Caves on Long Island and Salt Pond Hill Cave on Great Inagua
(Krieger 1937: 98). According to the Annual Report (Smithsonian
Institution 1938: 28-29), kitchen middens and burials were excavated
on Long Island, Great Inagua, and New Providence. The only conclu-
sions mentioned in either report are that data were uncovered point-
ing to a close cultural contact between the Lucayans and the Arawak
of Hispaniola, and that the tribal migrations of the Lucayans came
from Hispaniola, apparently at a relatively recent date (Smithsonian
Institution 1938: 29).

Mr. Krieger renewed his work in the Bahamas during January-May
1937. A brief, undetailed account of this expedition is given in
the Annual Report of the Smithsonian Institution for 1947 (Smithson-
ian Institution 1947: 16-17). Under a grant from Ernest N. May, Mr.
Krieger conducted an investigation of 15th century historic Indian
villages and some of the early Spanish settlements in the West Indies.
The expedition lasted from January 16th to May 5th, 1947, during
which time Mr. Krieger visited and made test excavations at Indian
villages referred to by Columbus in the journal of his first voyage
to the New World. One village site was located, and excavations
were conducted, near the town of Glinton (sic!), on the north end
of Long Island (Thompson 1949: 30). This site is presumably that
mentioned by Columbus (1893: 47) as having fields of excellent Indian

Mr. Krieger also reports excavation in a cave site with plain
pottery on the estate of the late Sir Harry Oakes on New Providence,
but no details were given about the nature of the site nor the speci-
mens recovered from it. No further information was available con-
cerning the expedition, although it has been assumed by the writer
that other village sites were among those investigated by Mr. Krieger.
Material from both of these surveys is at the United States National
Museum in Washington.

In July-August 1937 John M. Goggin, then an undergraduate student
at the University of New Mexico, made an anthropological reconnaissance
of Andros Island. He also studied and photographed the archeological


collection in the Nassau Public Library. Andros Island was covered
as thoroughly as possible over a four-week period, but no village
sites were located. The caves examined showed no signs of habitation,
but at Bain Hill, Mastic Point, on the northeast side of the
island, a single burial cave was located. It had been disturbed and
most of the bones taken away. The remaining bones were collected
and deposited at the University of New Mexico. Dr. Goggin secured
fourteen celts by purchase, but he found no ceramic specimens nor
other artifacts. Two of the celts are at the the University of New
Mexico, one is at the Yale Peabody Museum, and the rest are at the
University of Florida Laboratory of Anthropology. While in Nassau
he saw human bones in a cave near Lake Cunningham, about two miles
west of the city (Goggin 1937). An article in American Antiquity
(Goggin 1939) gives a report of this reconnaissance expedition.

During the summer of 1952 Dr. Goggin again visited the Bahamas
as a member of the University of Florida expedition to study the
economy and resources of the islands. He visited New Providence,
Exuma, Long Island, Fortune Island, Crooked Island, Rum Cay, Cat
Island, and Great Abaco. Although he inquired for "thunderbolts" or
celts everywhere he was able to purchase only one, on Rum Cay. This
celt is now at the University of Florida Laboratory of Anthropology.
Dr. Goggin visited the famous Hartford Cave petroglyph site near Port
Boyd on Rum Cay and a cave at Deadman's Bay on Long Island (Goggin
1952), but no artifacts were found in either of these caves. In
addition to these personal investigations he had reports of four
other sites, to be noted subsequently, and supplemented data from his
previous trip to the islands. Collections in Nassau, especially that
of Mrs. Hugh Johnson, were studied.

In 1954-55 Mr. Merrill Glenn, then a graduate student at the
University of Florida, reported celts from Mayaguana, exact proven-
ience unknown. No further data on these celts is available. In the
last two weeks of June 1955 the writer conducted a cursory survey of
New Providence and the Biminis. No new sites were located, though
the survey of New Providence was not as thorough as it might have
been. A reported burial site on the south end of North Bimini turned
out to be the local cemetery during the devastating cholera epidemics
of the late nineteenth century and early twentieth century.

During the month of April 1956 the writer and Dr. John Goggin,
at that time both with the Department of Anthropology at the Univer-
sity of Florida, made an exhaustive survey of Cay Sal. At that time
the late Howard Hughes was in the process of dredging a wide channel
from the central lagoon to the ocean and completing a runway large
enough to handle small commercial planes. In spite of this disrup-
tive project which was never completed few of the vegetation
areas of the island had been disturbed. No signs of any kind of
site were found on any part of the island, though a large, single
Spanish olive jar sherd was found protruding from a wall of one of
the channel cuts, approximately midway between the ocean and the
lagoon. This specimen reposes at the University of Florida Labora-
tory of Anthropology.


In 1957 and 1958 respectively, brief surveys were conducted
by the writer of Bimini and the west end of Grand Bahama as far
east as the then cleared site of the Lucayan Beach Hotel in Free-
port-Lucaya. These visits were made possible by small grants from
the Bryant Foundation of the Central Florida Museum through the
kindness of Fred Sleight and Junius Bryant. No sites were located
on either trip.

In 1960 Dr. Goggin visited San Salvador, at the request of Mrs.
Ruth Durlacher Wolper of the NewWorld Museum. At that time the
Palmetto site, later excavated by Charles Hoffman, was tested, and
the collection of artifacts gathered by Mrs. Wolper from various
sites on San Salvador was examined. Six test squares were dug at
the Palmetto Grove site, and smaller test excavations were carried
out at the Ward site, near Palmetto Grove, at the Bluff site, and
at the Cut Rock site (Hoffman 1967: 20).

In 1965 Charles Hoffman, then an instructor in the Department
of Anthropology at Florida Atlantic University in Boca Raton, exca-
vated the Palmetto Grove site (Hoffman 1967, 1970). This represented
the first thorough, documented excavation in the entire archipelago
and has largely formed the base for subsequent statments on Bahamian
ceramic styles to the present. During the same year Hoffman conducted
a site survey of Cat Island (Hoffman 1967: 19).

Beginning in 1965 a series of important surveys were completed
by Dr. William H. Sears of Florida Atlantic University and his
graduate students, particularly James C. MacLaury and Shaun D. Sulli-
van. The Central Bahamas region was the major area of focus of this
work, an exhaustive survey of Cat Island being undertaken from 1965
through 1967. MacLaury (1970: 27) followed this up with the digging
of stratigraphic test pits at five selected sites.

Between December 14th and December 29th, 1973, Shaun D. Sullivan
completed a site survey of Eleuthera, Harbour Island, and St. George's
Cay (Sullivan 1974: 1). In March of that same year the College Center
of the Finger Lakes of New York State, the working unit of a consor-
tium of private New York state colleges, initiated a program on San
Salvador which included, among other elements, prehistoric and his-
toric archeology. Marjorie K. Pratt was in charge of that segment
of the work. Several seasons of careful, productive work followed,
though at this writing in 1978, both continued follow-up as well as
the general ongoing program seem to have ceased, The initial exca-
vation was done at the Pigeon Creek site.

In 1976-78 Mr. Sullivan undertook a thorough site survey of
Middle Caicos, under the aegis of the Government of the Turks and
Caicos. This well-done survey was followed up by a thorough invest-
igation of a site given the identifying label MC-6 by Mr. Sullivan.
This is an important ceremonial site, complete with plazas and ball
court (Sullivan 1978?). It is to be hoped that a detailed present-
ation of the data from this site will soon be forthcoming.


An additional part of the Florida Atlantic University work
was a boat survey of both the Northern and Central Bahamas, con-
ducted by Sears from 1971 to 1973 (Sears and Sullivan 1978: 9).

In addition to the data provided by the surveys and excavations
discussed so far in this report, there is a considerable amount of
archeological material in private and museum collections, often
provenience unknown, both in this country as well as abroad. The
most important of these collections are the Godet-Greenway collec-
tion anda second smaller one at the Harvard Peabody Museum, the
Arnold collection at the Yale Peabody Museum, and the collection of
Lady Blake at the Museum of the American Indian.

The Godet-Greenway collection consists of 1,291 specimens,
Mainly ceramic, from the Bellevue site on North Caicos (listed by
the museum as coming from Grand or Middle Caicos, inasmuch as the
two islands - North and Middle Caicos - were formerly treated as
though they were a single island, Grand Caicos). The collection
was made between the years 1931 and 1936 by Mr. Godet of Bellvue,
and presumably came from the immediate vicinity of the settlement.
It was presented to Harvard through Dr. James C. Greenway. All the
material was collected from the surface. There are many decorated
sherds, and the collection has consequently proven of considerable
value in the delineation of ceramic styles.

The Arnold collection consists of forty-seven shell and stone
specimens collected by the late Mr. Benjamin Arnold of Albany, New
York. It was presented to Yale Peabody Museum around 1945-46 and
represents the work of Mr. Arnold over a period of thirty years.

In the collection are: 35 petaloid stone celts, 1 stone effigy
celt, 2 stone chisels, 1 irregular stone hammer-grinder, and 2 shell
celts. Only four of these specimens have any provenience, and for
these there is only island provenience - one petaloid celt from New
Providence, another from Rum Cay, and two from Andros. Published
reference to this collection is found in Moorehead's The Stone Age
in North America (1911).

Edith, Lady Blake, wife of Sir Henry Arthur Blake, governor of
the Bahamas from 1884 to 1887, gathered together a rather extensive
collection of artifacts and skeletal material, the majority of which
is now at the Museum of the American Indian.

The Public Library on Grand Turk has some archeological material
from the Caicos, exact island and site provenience unknown (Rainey
1940: 151: communication from C.B. Lewis, November 4, 1954). Two
duhos and two wooden platters or bowls seem to comprise the total

The Nassau Public Library has a collection of approximately 12
petaloid stone celts, 1 double-bitted stone celt, 1 stone chisel, a
single stone zemi, and several crania, all of the latter exhibiting
the usual parallelo-fronto-occiputal deformation, This material has


been cursorily mentioned by Rainey (1940: 149, 151; 1934: 7), Brooks
(1888: 216), Shattuck (1905: 421), and Goggin (1937 Field Notes,
Photographs; 1952 Field Notes).

Mr. Elgin J. Forsyth, former Commissioner of Andros, at one
time had a collection of eight celts from Mores Island (Goggin 1937
Field Notes).

The British Museum has a duho (CC1918-1) from Eleuthera. This
has been described by Joyce (1919: P;. 2, 3). Braunholtz (1951: 54-
55) subsequently describes the specimen again. It seems probable that
the Museum may have more material from the Bahamas, but lack of staff
has made it impossible to make any definite identification of such
specimens at the present time (communication from the British Museum,
December 1, 1953). Joyce (1916: 4) reports a monolithic ax, proven-
ience unknown, in the Museum's collection.

Mrs. Hugh Johnson of Nassau has a collection consisting of twenty-
four celts from Long Island, two of them double-bitted; a celt from
Cherokee Sound, Great Abaco; a celt from Mores Island; and a celt from
Andros. She also has two crania in her collection, both exhibiting
parallelo-fronto-occiputal deformation. They come from The Bogue on
Eleuthera and from Mayaguana (Goggin 1952 Field Notes).

Mrs. Herbert Brown, also of Nassau, has a unique petaloid stone
celt measuring 12 3/4 inches in length (Goggin 1952 Field Notes).
This is the largest celt known from the islands. The South Kensing-
ton Museum in London is reported to have been the recipient of a duho,
in turtle form, and some pottery from Black Bluff Cave, Rum Cay
(Goggin 1952 Field Notes), but it has not been possible to locate this
material (communication from Adrian Digby, November 10, 1954). A
single duho, analmost complete bowl, and some potsherds are in the
museum of St. John's University, Collegeville, Minnesota. They were
found by Father Arnold Mondloch, O.S.B., in a cave near Mortimer
settlement, Long Island (communication from Frederic U. Frey, May 10,
1954). No report has been published on this material.

A duho obtained from a family on Long Island by Father Mondloch
around 1940 is at St. Augustine's College in Nassau (communication
from Frederic U. Frey, May 10, 1954). Four crainia and a collection
of shell-tempered sherds were found at Smith Hill Cave, Andros Island,
by Captain L.W.B. Rees, who sent them to the American Museum of Nat-
ural History in New York (Goggin 1952 Field Notes). Dr. Harry
Shapiro, Curator of Physical Anthropology at the American Museum,
reported that he had no record of receipt of such material and that
all attempts to locate it were unsuccessful (communication from Harry
L. Shapiro, November 8, 1954). Another duho, from San Salvador, is
in the collection of the Academy of Natural Sciences in Philadelphia
(communication from Miss H. Newell Wardle to John M. Goggin, December
28, 1945), and a duho, reported by De Booy (1912: 87) is at the
United States National Museum in Washington. A monolithic ax is
reported from the Musee du Cinquantenaire, Brussels (L6ven 1935: 156;


Hamy 1906: Fig. 129). Site provenience, and even island provenience,
for this specimen is unknown.

The sum total of collections, surveys, and excavations in the
Bahamian archipelago is not, to date, impressive. It has been enough
to suggest various statments concerning the nature of Lucayan culture,
the origins of its characteristics in space and time, and the areal
distribution of its traits. What is clearly needed now is careful,
methodical excavation in all parts of the archipelago.

References Cited

Braunholtz, H.J.
1951 The Oldman Collection: Aztec Gong and Ancient Arawak

Stool. The British Museum Quarterly, xvi(2): 54-55.

Brooks, W.K.

1888 On the Lucayan Indians. Memoirs of the National Academy
of Sciences, 4(pt. 2): 215-223. Washington.

Columbus, Christopher (Colon, Cristobal)

1893 The Journal of Christopher Columbus (during his First

Voyage, 1492-93), and Documents Relating to the Voyages

of John Cabot and Gaspar Corte Real. Translated with
notes and Introduction by Clements R. Markham. The
Hakluyt Society, No. 86, London.

De Booy, Theodoor
1912 Lucayan Remains in the Caicos Islands. American Anthro-
pologist, new series, 14: 81-105.

1913 Lucayan Artifacts from the Bahamas. American Anthro-
pologist, new series, 15: 1-7.

Goggin, John M.

1937 Field Notes.

1939 An Anthropological Reconnaisance of Andros Island,
Bahamas. American Antiquity, 5: 21-26.

1952 Field Notes.

Granberry, Julian M.

1952 A Preliminary Ceramic Typology and Chronology for the

Bahamas. Manuscript on file at the Yale Peabody Museum,
New Haven.

1955 A Survey of Bahamian Archeology. Thesis on file at the

University of Florida Library, Gainesville, Florida.


Hamy, E.T.


The Cultural Position of the Bahamas in Caribbean
Archeology. American Antiquity, 22(2): 128-234.

Congress International d'Anthropologie et d'Archeologie,
Vol. 2, Monaco.

Hoffman, Charles A. Jr.
1967 Bahama Prehistory: Cultural Adaptations to an Island
Environment. Ph.D. Dissertation, University of Arizona,


Joyce, T.A.


The Palmetto Grove Site on San Salvador, Bahamas.
Contributions of the Florida State Museum. Social
Sciences, no. 16, pp. 1-26.

Central American and West Indian Archeology, New York.

Notes on a Wooden Stool from the Island of Eleuthera,
Bahamas. Man, 19: 1-2.

Kreiger, Herbert W.
1937 The Bahama Islands and their Prehistoric Population.
Smithsonian Institution, Explorations and Field Work,
1936, pp. 93-98, Washington.

Loven, Sven
1935 Origins of the Tainan Culture, West Indies. Elanders
Bokfryckeri Akfiebolag, Goteborg.

MacLaury, James C.
1970 Archeological Investigations on Cat Island, Bahamas.
Contributions of the Florida State Museum. Social
Sciences, no. 16, pp. 27-50.

Sears, William H. and Shaun D. Sullivan
1978 Bahamas Prehistory. American Antiquity,
43(1): 3-25.

Shattuck, George Burbank
1905 The Bahama Islands. The Geographical Society of
Baltimore, Macmillan Company, New York.

Smithsonian Institution
1938 Annual Report of the Board of Regents of the
Smithsonian Institution for the Year Ending
June 30, 1937. U.S. Government Printing Office,




Annual Report of the Board of Regents of the
Smithsonian Institution for the Year Ending
June 30, 1947. U.S. Government Printing Office,

Shaun D.
Archeological Reconnaisance of Eleuthera, Bahamas.
Thesis, Florida Atlantic University, Boca Raton,

1978? The Middle Caicos Project: An Overview. Undated
mimeographed report circulated at the 2nd Bahamas
Conference, Molloy College, October 1978.

Thompson, T.A.
1949 A Short Geography of the Bahamas, revised edition.
Nassau Daily Tribune, Nassau.


Irving Rouse

The first professional conference I ever attended, at Louisiana
State University in 1938, brought together the archeologists working
in the Midwestern and Southeastern United States. The two groups
had developed different conceptual approaches to archeology, and spent
much of their time debating the merits of their respective systems.
The Midwesterners, led by James B. Griffin, were primarily interested
in distinguishing "foci"-- the complexes, phases, or styles of Carib-
bean archeology--and they arranged them in chronological order so
as to provide the basis for solving problems of historical process,
such as migration, diffusion, and evolution. The Southeasterners,
led by James A. Ford, preferred to work in terms of individual types
of artifacts, which they used to set up local sequences of "periods."
They argued that periods were superior to foci for the purpose of
studying historical processes.

This disagreement about conceptual systems has continued to
bedevil us ever since, although it has not received so much attention
recently because present-day archeologists are more interested in
problems of adaptive process, which required the use of other kinds
of conceptual paradigms. Having myself been trained at Yale to use
the Midwestern system, I introduced it into the Caribbean area (Rouse,
1939). The Southeastern system was subsequently introduced into the
Lesser Antilles by Ripley P. Bullen (1963) and Charles A. Hoffman, Jr.
(1963), and into the Bahamas by James C. MacLaury (1970) and by William
H. Sears and Shaun O. Sullivan (1978). This set the stage for the
same kind of conflict between the two schools of thought that had
taken place in the United States.

Current usage of the term "series" is a case in point. In my
recent syntheses of Caribbean prehistory, I formulated a number of
series and noted that two of them, Meillacoid and Chicoid, appear to
have extended into the Bahamas (e.g., Rouse 1964, Fig 5). Sears and
Sullivan, in the article just cited, distinguish a third series in the
Bahamas, which they call Palmetto.

These series have figured prominently in the current debate about
the peopling of the Bahamas. Most participants in the debate have
treated all three series as comparable units and have assumed that
they are equally useful in tracing population movements. On the con-
trary, I employed the Midwestern system in formulating the Meillacoid
and Chicoid series, whereas Sears and Sullivan used the Southeastern
system in establishing the Palmetto series. Consequently, my units
are series of ceramic styles, whereas theirs is a series of ceramic

Styles, as the term implies, are indicative of population groups,


and so they may be used to trace the groups' movements. Types instead
express the ways in which each group was accustomed to make its arti-
facts. Consequently, types normally diffuse as customs from one group
to another, instead of being carried by the members of the first group
into the territory of the second group. This hardly qualifies them
for use in tracing population movements.

The fact that the West Indies extend like stepping stones between
North, Central, and South America has caused us to spend a dispropor-
tionate amount of time in studying population movements. As yet, how-
ever, we have reached a consensus only upon a single major migration,
that of the Saladoid peoples from South America. We succeeded in that
case because we worked in terms of a series of ceramic styles. Our
failure to reach agreement in other cases has been largely due to the
fact that we have been working instead in terms of individual types of
artifacts. The Palmetto series is a case in point.

Does this mean that the Midwestern approach is superior to the
Southeastern approach? Not at all. The two are complementary and
should be used in conjunction to obtain the best results. A comparison
of their respective procedures will illustrate this point.

1. In using the Midwestern approach, we focus upon sites rather
than specimens. We survey and excavate the sites in order to divide
them into components, each occupied by a single people. From every
component we obtain an assemblage, which consists of all the specimens
laid down by a particular people, that is, during a single occupation
of the site.

We proceed to classify the assemblages from a number of sites,
grouping together all those which look alike. We define each of the
resultant classes by noting the traits that are characteristic of its
constituent assemblages. The term "style" refers to the pattern of
diagnostic traits we obtain through this procedure.

As we do further survey and excavation, we use each style to
identify additional components occupied by the people it depicts. In
addition, we date the styles and, through them, the peoples they
symbolize, by setting up regional chronologies, such as the ones Allaire
and I have recently published (Rouse and Allaire, 1978).

Then we compare the styles occurring in adjacent parts of our
chronologies, that is, the styles possessed by peoples who immediately
succeeded each other or who lived side by side, and group together
those styles which show evidence of close relationship. In this way,
we arrive at series of styles and peoples (the two being opposite sides
of the same coin).

Finally, we are in a position to trace the movements of peoples.
We examine each series which has a horizontal or near-horizontal positic
on our chronological charts. If its constituent styles appear to form


part of a closely knit pattern of change and each is sharply distinct
from the style that preceded it, we may conclude that the series has
resulted from a population movement. If not, we must assume that the
series is due to some combination of the processes of diffusion and
independent invention, with only a minor amount of population movement.

2. In applying the Southeastern approach to our data, we focus
on specimens rather than sites, and classify them without regard for
their positions in the sites. We group together all the specimens
which look alike, thereby forming a class. This class is characterized
by a distinctive pattern of attributes that we call a "type." Alter-
natively, we may group the specimens in terms of their features rather
than as whole objects, in which case we obtain classes of features,
each with its distinctive pattern of attributes that we call a "mode."
(In this era of mechanical computation, many archeologists prefer to
work out the patterns by entering their attributes into computers.)

As additional specimens are collected, we use our types or modes
to assign them to classes, that is, we identify the new specimens as
whole or partial objects by determining which patterns of attributes
they possess. By applying the names for those patterns to the speci-
mens and their features, we are able to think of the specimens and
features conceptually instead of merely visualizing them.

We also use our types or modes as a means of setting up time
periods. For this purpose, we must confine ourselves to tightly
defined local areas, in order to eliminate geographical variations in
our types or modes and isolate their temporal qualities (Ford 1962).

When surveying and excavating sites for the purpose, we proceed
by components, but do not simply obtain an assemblage from each com-
ponent. Instead, we collect specimens from all possible points of
time within each component. We seek to obtain an adequate sample of
the types or modes present at each successive point of time.

We use the techniques of stratigraphy and seriation to arrange
the samples in their proper chronological order, and thereby work out
the patterns of change in the nature or frequency of our types or
modes. We divide the patterns into arbitrary periods, each of which
cuts across the flow of variations in the patterns (e.g., Rouse 1939,
Charts 5-7; Ford 1962, Figs. 6, 16).

Confusing the concepts of period and style, many archeologists
formulate periods which have the same order of magnitude as styles,
and set up regional chronologies of periods, as opposed to styles
(e.g., Bullen and Mattioni 1970). This is the source of the argument
as to the relative superiority of the Southeastern and Midwestern
approaches. It is a false argument, for it fails to recognize the
complementarity of the two approaches and, more particularly, the fact
that the Southeastern approach is concerned with temporal units
(periods), whereas the Midwestern approach is designed to study ethnic
units (peoples).


In my opinion, our temporal units should be made as fine as
possible. We should refine each period to the point where it covers
no more than a century. (Gary Vescelius informs me that he has
reached this point in the Virgin Islands by combining Rowe (1961)-
type stratigraphy and seriation with radiocarbon analysis and studies
of sedimentation rates.) Our sequences of periods would then become
effective dating systems, equivalent to Christian or native calendars.
Like the latter, they could be used to demarcate peoples' temporal
limits and to plot them more precisely in our regional chronologies.

But the usefulness of sequences of periods does not stop there.
They also make it possible to plot the distributions of individual
types or modes on regional chronologies and, by comparing them, to
distinguish series of types or modes. Just as series of styles are
the basis for inferring population movements from the archeological
evidence, so series of types or styles can be used to discern instances
of diffusion or independent invention.

Reverting to the problem of population movements within the
Bahamas, I have not seen all the recent collections and hence am not
in a position to make a definitive judgement. It is my impression,
however, that we do not yet have enough data to be able to carry the
Midwestern procedure through to the step of identifying known series
of styles and formulating new ones. Until we do, we cannot expect to
be able to resolve our arguments concerning the nature and direction
of the movements.

References Cited

Bullen, Ripley P.,
1963 Ceramic Periods of St. Thomas and St. John Islands,
Virgin Islands. William L. Bryant Foundation, Amer-
ican Studies, Report no. 4. Orlando.

Bullen, Ripley P. and Mario Mattioni,
1970 A Chronological Chart for the Lesser Antilles. Pro-
ceedings of the Third International Congress for the
Study of Pre-Columbian Cultures of the Lesser Antilles,
St. George's, Grenada, July 7-11, 1969, pp. 1-3.
Grenada National Museum, Grenada.

Ford, James A.,
1962 A Quantitative Method for Deriving Cultural Chronology.
Pan American Union, Technical Manual no. 1, Washington,

Hoffman, Charles A., Jr.
1963 Archeological Investigations on Antigua, West Indies.
M.A. thesis at the University of Florida, Gainesville.


MacLaury, James C.
1970 Archaeological Investigations on Cat Island, Bahamas.
Contributions of the Florida State Museum, Social
Sciences, no. 16, pp. 27-50. Gainesville.

Rouse, Irving

Rouse, Irving

Rouse, Irving

Rowe, John H.

Prehistory in Haiti, A Study in Method. Yale University
Publications in Anthropology, no. 21. New Haven.

Prehistory of the West Indies. Science,
144(3618): 499-513.

and Louis Allaire
Caribbean. In Chronologies in New World Archaeology,
edited by Clement W. Meighan and R.E. Taylor, pp. 431-
481. Seminar Press, New York.

Stratigraphy and Seriation. American Antiquity,
26(3): 324-330.

Sears, William H. and Shaun O. Sullivan
1978 Bahamas Prehistory. American Antiquity, 43(1): 3-25.


Victor A. Carbone

In his concluding remarks at the First Puerto Rican Symposium on
Archeology sponsored by the Fundacion Arqueologica, Anthropologica e
Historica de Puerto Rico, Willey (1976) briefly emphasized the need
for an ecological approach to Caribbean archeology. The papers by
Ranere, Linares, and Harris also served to underscore the need for a
systems oriented, cultural paleoecology in dealing with problems rela-
ted to the preceramic and the development of food producing economies.
In this paper I shall review the paleological evidence available from
the Caribbean area and discuss how these data may help to shed some
light on the changing picture of man-land relationships.

Three issues need to be considered. The first is a consideration
of the nature of the climatic changes which have occurred in the Cari-
bbean during the Holocene or recent times. The second involves the
biogeomorphic responses which these changes engendered. And, the third
relates to the cultural responses to changing environmental circum-
stances. The task, however, is not as simple as this three step pro-
cess implies for there are a number of theoretical questions which
need to be addressed involving contrasting interpretations of the paleo-
ecological record. Two alternative models of climatic change during
the Holocene have been proposed. One side postulates a continuous
warming trend from late glacial times climaxing in a "hypsithermal" or
warm interval in midpostglacial times followed by a gradual decline to
modern conditions (e.g., Deevy and Flint 1957). This model of post-
glacial climatic change was originally proposed by Antevs (1955)
and consisted of a three stage transition from the early postglacial
"anathermal" to the mid-postglacial "altithermal" and finally the late
postglacial "medithermal". The altithermal stage was presumed to have
been a period of maximum warmth and dryness extending from approximately
5000 to 3000 B.C. and it has been variously characterized as the "xero-
thermic interval" or "climatic optimum". The major problem with this
model has been its indiscriminate application over widely separated
geographical areas. Although climatic change may be considered a
global phenomenon, the changes do not assume the same character in
different locations. The model was originally developed based on
evidence from the Southwest and was later adopted and transported by
archeologists and other generalists to all parts of North and South
America without necessarily compensating for differences in geography.
As such it was found to be wanting, especially with the advent of more
sophisticated pollen analytical techniques which gave results that
differed markedly from expectations based on the model.

An alternative interpretation to this gradual model was provided
by Bryson (1970) and his co-workers. Their contention was that the
postglacial was characterized by abrupt climatic shifts, i.e., rapid,
step-like transitions in climatic regimes, from one quasi-steady state


to another. Thus, the paleoclimatologists' main task was to define
the character of the dominant pattern during each of the periods as
well as the time interval in which the pattern was operative (Bryson,
Barreis and Wendland 1970).

Recent work by Byson and others, not only in paleoclimatology
but also in palynology (Webb 1972) and fluvial geomorphology (Knox
1972, 1977; Carbone 1977) points to the episodic explanation as a
more satisfactory alternative. However, it should be emphasized that
acceptance of this model does not rule out the possibility of the
gradual model being correct, for the evidence indicates that the models
are not necessarily mutually exclusive and both may be plausible but
on different time scales. The precipitation data from Puerto Rico for
the last 100 years illustrates the short term episodic behavior of
climate in the area, showing the step-like transitions on a somewhat
smaller time scale (Hastenrath 1966).

The episodic model will therefore be adopted here for the inter-
pretations of the paleoenvironmental record of the Caribbean area and
it will be shown later how this model also serves as a guide for the
interpretations of the cultural record as well. The model assumes
that a square wave function accurately defines climatic behavior. The
various biogeomorphic responses which the climatic changes engendered
can then be represented by variants of this basic square wave by intro-
ducing factors such as lag into the system (Figure 1). In order to
lend substance to this idealized model, Wendland and Bryson (1974)
studied a set of global geologic-botanical data consisting of all C14
dates which marked discontinuities in pollen profiles, glacial records,
sea level changes and peat beds. These data were collectively analy-
zed for common discontinuities using statistical techniques and a
series of "environmentally significant change dates" were arrived at
which presumably define globally synchronous environmental episodes.
The nomenclature adopted for the episodes is that traditionally accepted
for the European postglacial sequence.

Now to explore the applicability of this model in a Caribbean
context. In recent years, the old notion that the lower latitudes,
especially the tropics, have experienced little environmental change
during the Quaternary has been replaced by a much more dynamic picture
of changing environments. Evidence presented by Thomas (1974), Fair-
bridge (1976), Burzer (1966, 1973), Selling (1948) and Vuilleumer (1971)
supports the interpretation of a dynamic tropical landscape during
the Pleistocene and Recent times supportive of both a long range grad-
ual model and short range episodic shifts. Fairbridge (1976) for
example, sees the tropical Holocene as having undergone three major

(a) the evolution from hyperaridity of the last pleni-
glacial to the postglacial pluvial;
(b) the "climatic optimum" of mid-postglacial times; and



(c) the post-optimum "deterioration that has involved des-
sication lakes, readvances of deserts, etc."
(Fairbridge 1976: abstract)

Selling, likewise, in discussing the late Quaternary history of
the Hawaiian vegetation as evidenced in the pollen diagrams from
mountainous regions, identifies three phases: an early period in
which the area of rainforest restricted (in comparison with present
conditions) in favor of drier vegetation types; a middle period in
which the area of rainforest expanded and a late period in which the
rain forest was replaced by drier vegetation types in parts of the
area occupied in period II. Superimposed on the long range pattern
is one consisting of more abrupt step-like transitions as idealized

Relative precipitation
\ !
| ! 1

Max 1 I I
Min |
1 1 i t
' ! \ 1 \
t ! i I I
' Relative vegetation !
I !
Max 1 I
I I ' 1 !
1 i I
I i t
i i 1
' i I i
I \ I ' 1
Min |
| 1 I i
I | | ! !
1 | Relative potential for |
| erosion |! !
1 i ' { 1
Max 1 i ! 1 1
I I 1
' I i
I 1 I
1 i I I
1 ! ! I
Min ' | }
I I | 1 I
P| ! | !
! | Relative sediment yield |
I ! I 1 1
Max I 1 ! ! !
I 1 i \ i
i 1 I 1 i
} I !
' I 1 | {
I I 1 i i
Min !
i i t

Figure 1. Idealized Square Wave Model
(from Knox 1972)


in the square wave model:

within the main periods (I-III) the development has
taken a complex undulating course. Waves of several orders
can be distinguished in the curves of the various rain forest
species as well as in the shifts of the boundaries between
the rainforests and the drier forest types (Selling

Table 1. Dates of Significant Environmental Change

Episode Major discontinuity Minor discontinuity
(BP) (BP)

sub-Boreal 4240
Atlantic 6050
8,490 7740
late Glacial

Selling continues his description of the "wave-like rhythm" and "sym-
metrical arrangement of the waves" although he cautions that the sym-
metry is not total:

.* it would seem that the great changes have been brought more
or less by leaps, though not without forerunners, and that they
have ended with a more gradual dying out. The conditions charac-
terizing period III are considered to have come more or less
like a climatic shock. (Selling 1948:138)

With specific reference to tropical South America, Vuileumer's
conclusion (1972) was as follows:

intensive work on lowland tropical taxa and. on montane
forest elements now justify the conclusion that the floras and
faunas of these areas were greatly affected by Pleistocene cli-
matic shifts. In the broad region of South America that lies
within the tropics, a series of humid-arid cycles drastically
and repeatedly altered vegetation patterns during the Quaternary.
Both montane and lowland rainforests were fragmented during dry



periods and were able to reexpand during humid phases. .

On a more generalized level, climatic oscillations may be traced
in cores taken from the bed of the equatorial and North Atlantic
oceans as well as the Caribbean (Wollin, G., D.B. Ericson and M.
Ewing 1971). Change in past temperatures may be traced by studying
changes in assemblages of temperature sensitive planktonic forammini-
feral deposits, by changes in ratios of 018/016 in specific species
of foramminifera and also by measuring rates of calcium carbonate
sedimentation (CaCO3) (Wiseman 1966). Wiseman (1966:84) studied the
CaCO3 sedimentation rates from deep sea cores and concluded that:

In the Equatorial Atlantic core short-period oscillations are
superimposed on long period trends, and there would seem to be
a close correlation between the dates of these changes and con-
tinental climatic oscillations found in higher latitudes.

Wollin et al. have generated a series of climatic curves based
on changes in foramminiferal assemblages including variations in the
abundance of the Globigerina menardii group, the coiling ratios of
Globigerina pachyderma (left coiling indicates cold temperatures,
right coiling-warm temperatures) and Globorotalia truncatulinoides,
all of which are in generally good agreement. In these curves there
is evidence for approximately 20 short climatic oscillations exhibiting
a 550 year periodicity, superimposed on the long term postglacial trend

The evidence supportive of the episodic nature of environmental
change in the Caribbean takes a number of forms. Hastenrath (1966)
in evaluating these data from the Central American area briefly indi-
cated the potential effects of climatic fluctuations on hydrology,
plant life and ultimately human activity, and although he does not
admit the existence of periodicities in the recent data he does accept
the fact that large scale climatic fluctuations have occurred and are
responsible for much environmental change. Other more concrete data
bearing on the total Holocene paleoenvironmental picture in the Carib-
bean is derived from a number of sources including pollen profiles
from Columbia (van der Hammen 1961, 1962, 1966; van der Hammen and
Gonzales 1960a, 1960b; van der Hammen, Gonzales and Flint 1966;
Wijmstra and van der Hammen 1966; Gonzales and van der Hammen 1966;
van Geel and van der Hammen 1973), Panama (Bartlett 1967; Bartlett
and Barghoorn 1972), Costa Rica (Martin 1964), Guatemala (Tsukada
1966; Vaughn 1978), British Guiana (van der Hammen 1961), Puerto
Rico (Ogle 1971), Venezuela (Muller 1959; Leyden, et al. 1978).
Pedologic and other geomorphic studies are also contributing informa-
tion on the late Quaternary sequences. These include studies of paleo-
sols in Colombia (Folster and Hetsch 1978; Folster, Hetsch and
Schrimpft 1977), mineralogy of deep sea cores from the Equatorial
Atlantic (Damuth and Fairbridge 1970); coastline changes and progra-
dation sequences in Puerto Rico (Kaye 1959a) and studies of cave
deposits on Mona Island (Kaye 1959). The general distribution of
these paleoecological study sites is shown in Figure 2.

The work in northern South America is obviously the most exten-
sive. Beginning in the early sixties van der Hammen and his associates


1. Colombian sequences (van der Hammen, et al.)
2. Lago de Valencia, Venezuela (Leyden 1978)
3. Orinoco delta sediments (Muller 1959)
4. British Guiana coastal sediments (van der
Hammen 1961)
5. Equatorial Atlantic sites (Wollin, et al.
1971; Wiseman 1966)
6. Isla Mona (Kaye 1959b)
7. Puerto Rico shorelines (Kaye 1959a)
Figure 2. Paleoecological

8. Luquillo Forest (Ogle 1971)
9. North Atlantic deep sea cores
10. Lake Quexil, Guatemala (Tsukada 1966;
Vaughn 1978)
11. Costa Rica (Martin 1964)
12. Panama Canal Zone (Bartlett 1969)
13. Caribbean deep sea cores (Wollin, et al.1971)
14. Equatorial Atlantic mineralogy (Damuth
and Fairbridge 1970)
Data Site Locations


published a series of papers dealing with the late Quaternary vege-
tational and paleoclimatic history of this area in which they indica-
ted their possible correlation with the European late glacial and post-
glacial sequences. Two of the major conclusions from these studies
were as follows:

(1) On the basis of the temperature curves for the Upper Pleisto-
cene and Holocene and combined radiocarbon dates, it may be said
that the glacials and interglacials and also the interstadials
and probably the minor climatic fluctuations of the Holocene of
Equatorial South America are perfectly contemporaneous and may
directly be correlated with the ones known from the temperate
regions of the northern hemisphere.

(2) In the late glacial and Holocene of the Sabana de Bogota
eight zones may be distinguished. In so far as C14 dates are
available, the zone limits seem to coincide very well with the
European zone limits. The succession of dry-wet-dry-wet during
the Holocene in the Sabana de Bogota seems to correspond very
well with the European Boreal-Atlantic-Subboreal-Subatlantic
succession. (van der Hammen and Gonzales 1960)

The data from other parts of northern South America are not as
clear cut. However, the sequences derived from coastal sediments in
British Guiana (van der Hammen 1961) and the Orinoco delta (Muller,
1959) do show the drastic environmental changes which have taken place.
These sequences are complicated by the nature of postglacial sea level
fluctuations. But, regardless of whether the record is interpreted
as representing transgressive/regressive effects, or as others (e.g.,
Jelgersma 1966) have suggested, differences in the amounts of sediment
transport of the rivers, either interpretation is suggestive of cli-
matic changes.

Data from Venezuela (Leyden, et al. 1978) at the annual meeting
of the Ecological Society of America in Athens, Georgia, also supports
an oscillating picture of Holocene environments. Analyses of the
sediments at Lago de Valencia produced palynological, chemical and
microfaunal evidence which attests to alternating wet-dry cycles in
the basin. Changes in the pollen spectrum suggestive of periods al-
ternating between humid and savanna types of vegetation seem to cor-
respond with changes in the sedimentary stratigraphy as well as changes
in the total rates of pollen influx. Changes in the fossil pigment
data (Chlorphyll and carotenoids) as well as in the microfaunal record
are also supportive of the inferred oscillating pattern. Leyden, et
el. conclude that "significant fluctuations in both watershed vegeta-
tion and in the lake hydrological system have occurred during the last
10,000 years" and that the "fluctuations appear to be rapid and dra-

One of the more fascinating asides to this project is their sug-
gestion that the highly significant increases in the pollen of the
Chenopod/Amaranth family which is experienced at 7200 B.P. is indica-
tive of incipient cultivation.




The data from Panama are summarized in Table II and, although
not as neatly resolved as the Colombian sequences, the general indi-
cation is that the sequence follows the same pattern. The question
of vegetational structure and its possible paleocultural significance
will be taken up in a later section.

Table II.

Panama Canal Zone Paleoenvironmental
Sequence (Source: Bartlett 1972)

I______________________ I

Fresh water swamp fully de-
veloped; strong herbaceous
swamp element; sedges and grasses
increase; aboreal pollen decrea-
ses and nearly disappears towards
the top; maize and manioc pollen
present; many charcoal fragments.


4200 B.P.

Transition from mangrove to
freshwater swamp; Rhizopora dis-
appears; sedges and grasses
abound; Myrica and Ilex present;
wild maize pollen.

1 1-

7300 B.P.

Rhizopora abundant; mangrove
swamp; high altitude species
decline in abundance and dis-
appear towards the top; Urti-
cales pollen continues in
high percentages; "chayote"
pollen possibly present.

I I-

9600 B.P.

11300 B.P.

Rain forest genera; Rhizopora
and other swamp species rare
or absent; several high alti-
tude species present indicat-
ing range extension; Alchornea
pollen abundant; Urticales pol-
len represents vegetation char-
acteristic of disturbed ground.

I i


Return to more humid
conditions; open
water; slash and burn
agriculture affects
vegetation; defor-

Different vegeta-
tional structure sug-
gested no modern
analogs; climatic
change indicated -
very dry; season-
ality more pro-
nounced on Atlantic

Gradual temperature
rise indicated by
disappearance of
high altitude species.

Temperature approx-
imately 2.50 colder;
vegetational struc-
ture has no modern

Interglacial; high
sea level.






Figure 3. Coastal geology, Punta Vacia Talega to Punta Miquillo


Evidence for changes in the Puerto Rican landscape during late
Quaternary and recent times comes from coastal progradation sequences.
The changes in coastal features are reflective not only of changing
sea levels but also of shifting climatic regimes since the rates of
progradation are going to be dependent to a certain extent on the
amount of sand brought down by the rivers and this sediment load is
going to be related to climatic parameters such as precipitation.
Stages in sea level rise as well as possible climatic shifts result-
ing in changing sediment load are especially evident in the prograda-
tion sequence on the north coast between Punta Vacia Talega near the
mouth of the Rio Grande de Loiza and Punta Miquillo at the mouth of
the Rio Espiritu Santo. This area shows the most extensive develop-
ment of fossil beach ridges and ancient shorelines on the island.
Other sequences are found at Ensenada de Boca Vieja on the north and
Punta Cabullon near la Playa de Ponce. (The sequence between Punta
Vacia Talega and Punta Miquillo is shown in Figure 3). An example of
changes involving both sea level rise and possible climatic change
can be seen in the transition from the marsh environment to the
south and the prominent beach ridges to the north, the transition
itself signalling a shift (Kaye 1959a).

Kaye (1959a) presented an estimation of the age of beginning of
development of these coastal sequences. However, his estimates to
a large degree were based on a presumed age for a preceramic site of
A.D. 890. This is obviously considerably off the mark and Kaye's
sequence is therefore in need of revision. But, the important thing
to remember is the fact that great changes in the coastal landscapes
have taken place for at least the last five to six thousand years
and these changes have great significance in relationship to archaic
settlement and subsistence patterns since the changes in shorelines
would have considerably affected the setting and resource availability
at sites such as Cayo Cofresi (Veloz Maggiolo, et al. 1975) as well
as other archaic shell collecting stations. The changing stages of
sea level rise are shown in Table III as indicated by Kaye's data but
it should be kept in mind that these dates will have to be revised on
the basis of what we now know about the dates for preceramic sites
in Puerto Rico.

Another major source of evidence for a series of humid/arid
cycles in Puerto Rico and the northern fringe of the Caribbean is the
depositional sequence in the caverns of Mona Island. The caverns
themselves seem to be the result of solution during a Pleistocene
high sea level stand, perhaps a mid-pleistocene interglacial. Cavern
formation was followed by a period during late Pleistocene times in
which bats inhabited the caves and were responsible for large accumu-
lations of guano. Guano deposition seems to have been accompanied
and followed by a humid period which resulted in intense leaching of
the guano deposits. The guano deposits were subsequently capped by
a pavement of dripstone, indicating a shift to drier conditions.
According to Kaye (1959b), dripstone formation was "possibly brought
about by a rather abrupt change in climate to greater aridity (since)
humid caves inhibit evaporation and therefore the formation of drip-
stone". Just when in the Quaternary the change occurred is of course
not known but the suggestion is offered by Kaye that it may have been
at the close of the postglacial warm interval, the so-called "hypsi-



Further details of the characteristics of paleoclimates on Mona
Island are derived from a mineralogical study of the guano deposits
themselves. The deposits contain both the hydrated and non-hydrated
forms of CaHPO4, known as brushite and monetite respectively. Brush-
ite is known to be formed at low temperatures while monetite forms
at higher temperatures. Specifically, brushite is known to form at
temperatures below 71 F (250C) (Kaye 1959b). This temperature is
somewhat lower than surface and cavern temperatures reported from
Mona, thus, brushite formation reflects a somewhat colder paleocliamte.
On the other hand, the presence of monetite indicates a climate simi-
lar to or possibly warmer than today's.

It should be emphasized that if the dating of the dripstone pave-
ment is indeed amid to late postglacial event then perhaps the best
source of more recent paleoclimatic data for the area is yet to be
tapped namely the internal accretionary structure of the dripstone
stalagmites in the Caverns of Mona Island. Patterns of dripstone
accretion are dependent on ground water/precipitation and therefore
should reflect shifts in climatic regimes.

One last significant climatic episode on Mona Island emerges from
the historical accounts (Kaye 1959b). The records contain a number
of references to the fertility of Mona Island during the late 15th,
16th, and 17th centuries, as well as to its use as a provisioning and
watering station. This represents a climate on the island considerably
more humid and possibly cooler than today and may indicate a possible
correlation with a climatic episode known as the "Little Ice Age", a
phenomenon for which there is substantial evidence across the temper-
ate latitudes and especially in Europe. The total late Quaternary

Table III. Generalized Sea Level Changes in
Puerto Rico (from Kaye 1959)



5000 ----------------

7000 -

8000 I


climatic picture for Mona Island is presented in Table IV as infer-
red from the data presented by Kaye. Needless to say, I would agree
with Kaye when he states that "there are good reasons to think that
climatic fluctuations occurred at Isla Mona during the Pleistocene
(and I would add Recent) varying from climates more humid to climates
more arid that that of today."

Table IV.

Inferred Paleoclimatic Sequence from Mona Island
(Source: Kaye 1959)

One last bit of evidence which somewhat obliquely sheds some
light on the paleoenvironmental picture in Puerto Rico is the pollen
work carried out by Ogle (1971) in the mountain forests of El Yunque.
Three pollen diagrams are available from Sphagnum bogs in the Luquillo
forest. The analyses of these bogs is somewhat less than complete and
no C14 dates are available. However, the presence of Casuarina pollen
(australian pine) indicates a recent date for the bogs since these
pines were not introduced into Puerto Rico until the 1920's (Ogle
1971). The existence of the bogs themselves may be an indicator of
some recent climatic changes. The diagrams indicate that potentially
very productive pollen sites do exist on the island and a pollen analy-
tical study of these and other sites should be encouraged.

To close the circle around the Caribbean, and by way of introduc-
tion to the next section of this paper, brief mention should be made
of the results reported by Vaughn (1978) at the recent Ecological Soc-
iety of America meeting on analyses of pollen data from late sites in
the Guatemala Mayan area. Although Vaughn's investigations are focus-


1940 Recent Arid, Warm

1500- Little Ice Age Humid, cooler
1800 (Cultivation on the
island; ground water

3-4000B.P. Late post- Arid, warmer

6-8000B.P. Mid-postglacial Humid, warm (Guano
deposition; leach-

8-12000 B.P. Late glacial/ Humid, cooler
early postglacial (Guano deposition)

20000 B.P. Mid to late Cavern formation
Pleistocene (dissolution during
high sea level?)




ing primarily on the changes attendant with the rise and fall of the
classic civilizations in Guatemala, his results from Lake Quexil
leave little doubt of the influence of climatic factors in the evol-
ution of this complex sociopolitical system. Changes in the patterns
of seasonality and especially rainfall are evident in the data, and
it is interesting to note that the period exhibiting the most wildly
oscillating climatic behavior is associated with the early classic

The above review of the paleoecological evidence from the Carib-
bean area should be sufficient to indicate that the climate and en-
vironment in the area have undergone considerable change, and that
these changes have generally assumed an episodic character as ideal-
ized in Bryson's square wave model. In an effort to explore their
possible paleocultural significance it should be possible to postulate
a similar square wave function variant as was done with the various
environmental variables. This variant would represent the cultural
responses to the changing environmental conditions.

In order to discover the possible correlations between environ-
mental and cultural discontinuities, Wendland and Bryson (1974) under-
took an analysis on a global scale of the different cultural manifesta-
tions and the radiocarbon dates available for them, on the assumption

S. the cultural history of man may be a proxy record of
environmental history. This is not to suggest that man fully
or solely responds to his environment and its changes, but
rather that initial conditions may be established by the
environment. If this concept of the relationship between
man and the environment is even partially correct one would
expect that cultures, once established would tend to continue
throughout one or more times of stable environment, but be
more likely to terminate or change at climatic discontinuities
rather than during stable intervals. thus the most likely
times of beginning and ending of climatically sensitive con-
tinua such as cultural regimes, should be related to the times
of climatic discontinuity (Wendland and Bryson 1974:14)

Through a statistical analysis of radiocarbon dates the authors
were able to establish major cultural episodes as defined by discon-
tinuities in the data matrix. A table comparing the discontinuities
in both the environmental and cultural records is presented here in
Table V, taken from Wendland and Bryson, 1974. The table shows very
close agreement between the two sets of data.

An attempt to explore this culture/environment relation-
ship can also be made from a theoretical point of view. Clarke (1969)
explored the dynamic equilibrium between the sociocultural system (S)
and the environmental system (E) by expressing the relationship as an
equality in which the two systems are related through some intermed-
iary, k, which serves as a "buffer" and has a regulatory or insulatory
capacity (Clarke 1969:126). This he expressed as follows:

E = (k)S eq. (1)


The insulatory or regulatory capacity of a cultural system is propor-
tional to its variety so that from the anthropological point of view
the system's ability to regulate or insulate itself is going to be
dependent on the relative variety in artifacts and assemblages. Var-
iety, then is a crude measure of k and is the major parameter which
can be utilized to define the cultural function if one is to explore
the cultural responses to environmental change. Clarke (1969) also
showed how artifacts, types, assemblages, etc., exhibit a systematic
ontogeny in a pattern analogous to birth, maturity and death, with
each state possessing limiting threshholds. He characterized this
ontogeny in a five step manner:
Phase (0) preformative threshold type Sl/So birth/death
Phase (1) formative growth
Phase (2) coherent type S1 maturity
Phase (3) post coherent decline
Phase (0) preformative threshhold- type Sl/S2 death/birth

Table V.

Comparison of Dates of Discontinuity Determined
From the Environmental Record and From the
Cultural Records (Yr BP). Cultural Dates
Stratified by Significance Values, "t"



Botanic Cultural
Major Minor >50 50-30 30-10



SB III 3,110





A IV 5,680












Each of these states is characterized by a series of statistical
trends or constraints.

If equation (1) is correct then Clarke's model of material cul-
ture system ontogeny, utilizing variation as the major parameter, can
be placed within a framework of environmental change embodied by the
square wave model. The changing cultural function will be expres-
sive of variation within the material culture subsystem and this vari-
ation is going to follow the outline of Clarke's ontogenetic scheme.
This is shown in Figure 4.

Climate--Relative Precipitation

Relative Vegetation

IS S1 S2 S Type State
I 1 I I


Material Culture Variation

Figure 4. Changing Cultural Function (Variability) in
The Context of Idealized Square Wave Model.

f=formative; pc=post coherent)
Material Culture Variation

The Context of Idealized Square Wave Model.


The model's inherent predictive capability can thus become very
useful in cases where sufficient environmental data are available
since periods of possible environmental stress can be defined and
these should have occasioned the sharpest cultural repercussions.
Meggers (1971) has suggested an example of the cultural/environment
relationship in the Amazonian setting. She suggests that the wide
dissemination of certain language families within the Tupi-Guarinian
and Arawakan stock which have been dated through lexicostatistical
techniques coincides with the late postglacial "xerothermic interval"
experienced in Amazonia. According to Meggers:

Both stocks are assigned a homeland in southwestern Amazonia
and the adjacent Andean Highlands. Both contain one family
that became widely disseminated over the eastern lowlands,
and the dates estimated for both of these migrations fall
within the most recent period of aridity (4000 B.P.). .
Meggers 1971:36)

Lathrap's (1970) model of population dispersal which is essentially
based on similar linguistic evidence can likewise be fitted into the
episodic environmental scheme and its implication with respect to
migrations into the Antilles should not be lost.

Another aspect of this review which significantly affects esta-
blished notions of the paleocultural record is the evidence on chang-
ing environmental structure. The data where available in substantial
detail (e.g., northern South America and Panama) indicate that drastic
changes have occurred during the postglacial and that the conditions
obtained during the various episodes have no modern analogs. Inter-
pretations of the paleoecological record which assume a correspondence
between modern and "fossil" biotic communities and a simple zonal dis-
placement of community elements are untenable (e.g., Hester's recon-
struction of late Pleistocene environments in South America on the
basis of zonal altitudinal displacements (Hester 1966). The record
is much more complex. In the temperate areas the picture emerging
for late glacial and early postglacial environments is one of an eco-
logical mosaic or patchwork of environments with an intermingling of
elements and habitats which do not coexist today (Carbone 1976). The
same holds true for the mid-postglacial. The pollen evidence from
Panama, for example, shows a different environmental structure than
that which prevails today. Interpretations of the cultural data (e.g.,
Ranere 1976) which do not take into account the different setting
vis-a-vis settlement and subsistence patterns will need to be re-
evaluated in the light of changed environmental circumstance.

Finally, I would like to close with a call for a much more con-
certed effort at dealing with what have recently been called "ecofacts".
It is this ecological debris, including floral elements such as pollen
and phytoliths (Carbone 1977), charcoal and wood fragments, macro-
and micro-faunal remains such as snails, diatoms, etc., as well as
inorganic and mineralogical remains in sedimentary sequences and the
pedological and geomorphological sequences themselves which hold many
of the clues necessary for unravelling the changing picture of man-
land relationships in the Caribbean.



References Cited

Antevs, E.
1955 Geologic-climate dating in the West. American
Antiquity 20:317-335.

Bartlett, A.S.
1967 Palynological Studies of the Gatun Basin, Panama. A
History of the Vegetation, Climate and Sea Level Change
During the Past 12,000 Years. Unpublished Ph.D. Dis-
sertation, Dept. of Biology, Harvard University.

Bartlett, A.S. and E.S. Barghoorn
1972 Phytogeographic History of the Isthmus of Panama During
the Past 12,000 Years (A History of Vegetation, Climate
and Sea Level Change.) In Vegetation and Vegetational
History of Northern Latin America. Alan Graham, Editor.

Bryson, R.A.

Bryson, R.A.,

Butzer, K.W.


Carbone, V.A.


Clarke, D.

The Character of Climatic Change and the End of the
Pleistocene. In Abstracts of the First Meeting of the
American Quaternary Association (AMQUA). Bozeman,

D.A. Barreis, and W.M. Wendland
The Character of the Late Glacial and Postglacial Changes.
In Pleistocene and Recent Environments of the Central
Great Plains. edited by W. Dort, Jr. and J.K. Jones, Jr.
pp. 53-74. University of Kansas Press, Lawrence.

Climatic Changes in the Arid Zones of Africa During Early
to Mid-Holocene Times. In World Climate from 8000 to
0 B.C. J.S. Sawyer, Editor. Royal Meterological Society,

Environment and Archaeology, Aldine, Chicago.

Environment and Prehistory in the Shenandoah Valley. Un-
published Ph.D. Dissertation, Department of Anthropology,
Catholic University, Washington, D.C.

Phytoliths as Paleoecological Indicators. In Amerinds
and Their Paleoenvironments in the Northeast, Newman,
W.S. and B. Salwen, Editors. Annals of the N.Y. Academy
of Science, Vol. 288, pp. 194-205.

Analytical Archaeology. Metheun and Co. London.

Damuth, John E. and R.W. Fairbridge
1970 Equatorial Atlantic Deep Sea Arkosic Sands and Ice-age
Aridity in Tropical South America. Geological Society
of America Bulletin, 81: 189-206.



Deevey,E.S. and R.F. Flint
1957 Postglacial Hypsitherman Interval. Science, 125:

Fairbridge, R.W.
1976 Effects of Holocene Climatic Change on Some Tropical
Geomorphic Processes. Quaternary Research, 6:529-556.

Folster, H. and W. Hetsch
1978 Paleosol Sequences in the Eastern Cordillera of Col-
ombia. Quaternary Research, 9:238-248.

Folster, H., W. Hetsch and Schrimpff, E.
1977 Late Quaternary Paleosols in the Western and Central
Cordillera of Colombia. Paleogeography, Paleoclima-
tology, and Paleoecology. 21:245-264.
Gonzales, E., T. van der Hammen and R.F. Flint
1966 Late Quaternary Glacial and Vegetational Sequence in
Valle de Lagunillas, Sierra Nevada del Cocuy, Colombia.
Leidse Geologische Mededelingen 32:157-182.

Harris, Peter
1976 The Preceramic Period in Trinicad. Proc. of the First

Puerto Rician Symposium on Archaeology. Robinson,
Linda. S., editor, FAAH Informe Numero 1. San Juan.

Hastenrath, S.
1966 Recent Climatic Fluctuations in the Central American
Area and Some Geo-Ecological Effects. In "Colloquium

Geografium: vol. 9, Geo-Ecology of the Mountainous
Regions of the Tropical Americas, pp. 187-194. Bonn.

Hester, J.
1966 Late Plesitocene Environments and Early Man in South
America. The American Naturalist, 100:377-388.

Jelgersma, S.
1966 Sea Level Changes During the Last 10,000 Years. In
World Climate from 8000 B.C. to 0 B.C., J.S. Sawyer,
Editor. Royal Meteorological Society, London.

Kaye, Clifford A.
1959a Shoreline Features and Quaternary Shoreline Changes

Puerto Rico. U.S. Geological Survey Professional
Paper 317-B. GPO Washington, D.C.

1959b Geology of Isla Mona Puerto Rico and Notes on Age of

Mona Passage. U.S. Geological Survey Professional
Paper 317-C. GPO, Washington, D.C.

Lathrap. D.
1970 The Upper Amazon. Thames and Hudson. London.


Leyden, Barbara, D.R. Whitehead and D.G. Frey
1978 A Palynological Investigation of Holocene Sediments
From Lake Valencia, Venezuela. Paper presented at the
1978 meeting of the Ecological Society of America.

Martin, P.S.


Effect of Pleistocene Climatic Change on Biotic Zones
Near the Equator. Yearbook, American Phiosophical
Society, 1960:265-267.

Paleoclimatology and a Tropical Pollen Profile.
Proceedings of the 6th Congress of the International
Association for Quaternary Research. Warsaw. 2:319-323.

Meggers, B.
1971 Amazonia: Man and Culture in a Counterfeit Paradise.
AHM Publishing Co. Arlington Heights, Illinois.

Muller, J.
1959 Palynology of Recent Orinoco Delta and Shelf Sediments.
Micropaleontology, 5:1-32.

Ogle, Carol J.
1971 Pollen Analysis of Selected Sphagnum-Bog Sites in Puerto
Rico. In A Tropical Rain Forest, E. Odum, Editor.

Ranere, A.
1976 The Preceramic of Panama; The View From the Interior.
Proc. of The First Puerto Rican Symposium on Archae-
ology. FAAH Informe Numero 1, San Juan.

Schobinger, Juan
1969 Prehistoria de Suramerica. Nueva Coleccion labor.
Editorial Labor, S.A. Barcelona.

Selling, O.
1948 Studies in Hawaiian Pollen Statistics, III. On the
Quaternary History of the Hawaiian Vegetation. Bernice
P. Bishop Museum Special Publication No. 39. Honolulu.

Thomas, Michael
1974 T

ropical Geomorphology.

Tsukada, M.

The Pollen Sequence. In U.M. Cowgill, et al. The
History of Laguna de Petenxill. Memoirs of the
Connecticut Academy of Arts and Sciences, 17:1266pp.

van der Hammen, T.
1961a The Quaternary Climatic Changes of Northern South
America. Annals of the New York Academy of Science,



1961b A Palynological Study on the Quaternary of British
Guiana. Leidse Geologische Mededelingen, 29:125-180.

1962 Palinologia de la region de "laguna de los Bobos".
Historia de su clima, vegetacion y agriculture
durante los ultimos 5000 anos. Acad. Colombiana de
Ciencias Exactas, Fisicas y Naturales. Revista,
v. 11, pp. 359-361.

1966 Climatic and Vegetational Succession in the Equatorial
Andes of Colombia. In "Colloquium Geograficum", vol
9, Geoecology of the Mountainous Regions of the
Tropical Americas. pp. 187-194. Bonn.

and Gonzales, F.
1960a Upper Pleistocene and Holocene Climate and Vegetation
of the Sabana de Bogota. Leidse Geologische Medede-
lingen, 24:261-315.

1960b Holocene and Late Glacial Climate and Vegetation Paramo
de Palacio. Geologie en Mijnbouw, 22:737-746.

van Geel, B. and van der Hammen, T.
1973 Upper Quaternary Vegetational and Climatic Sequence of
The Fuquene Area (Eastern Cordillera, Colombia) Paleo-
geography, Paleoclimatology and Paleoecology, Vol. 15.

Vaughn, Hague H.
1978 An Absolute Pollen Diagram from Lake Quexil, Peten,
Guatemala. Paper presented at the 1978 Annual Meeting
of the Ecological Society of America. Athens, Georgia.

Veloz Maggiolo, M., J.C. Gonzalez, E.J. Maiz and E. Questell Rodriguez
1975 Cayo Cofresi: Un Sitio Preceramico de Puerto Rico.
Ediciones de Taller. Santo Domingo.

Vuillemier, B.S.
1971 Pleistocene Changes in the Fauna and Flora of South
America. Science, 173:711-780.

Webb, T. and R.A. Bryson
1972 Late and Postglacial Climatic Change in the Northern
Midwest U.S.A.: Quantitative Estimates Derived From
Fossil Pollen Spectra by Multivariate Statistical
Analysis. Quaternary Research, 2:70-115.

Wendland, W. and R.A. Bryson
1974 Dating Climatic Episodes of the Holocene. Quaternary
Research, 2:9-24.

Wijmstra, T.A. and T. van der Hammen
1966 Palynological Data on the History of Tropical Savannas
in Northern South America. Leidse Geologische



Wiseman, J.D.H.
1966 Evidence for Recent Climatic Changes in Cores From
the Ocean Bed. In World Climate from 8000 to 0 B.C.
J.S. Sawyer, Editor. Royal Meteorological Society,

Wollin, G., D.B. Ericson and M. Ewing
1971 Late Pleistocene Climates Recorded in Atlantic and
Pacific Deep-Sea Sediments. In Late Cenozoic Glacial
Ages, Karl Turekian, Editor. Yale University Press.
New Haven.


Shaun D. Sullivan


This writer in conjunction with forty-one volunteers and with the
financial support of the University of Illinois and Earthwatch has
conducted ten months of archeological surveys and test excavations in
the Turks and Caicos Islands from August, 1976 through August, 1978.
This paper is intended to provide a broad outline of the results of
those investigations and to anticipate some of the conclusions of more
detailed analysis of the data which will be the subject of subsequent


% zone 2

T.&C. Is.
Zone 1

Figure 1. Northwestern Caribbean



The archeological materials recovered have yielded data which
appears to bear upon several intriguing questions concerning the
processes of Arawakan exploitation and colonization of new island
territories. Although this new body of information strongly indicates
that the prehistory of this island group differs significantly from
that of the Bahamas proper, it also confirms that the Bahamas and
Turks and Caicos Islands cannot be adequately studied in isolation
from each other.

The Turks and Caicos are located at the extreme southeast end
of Bahama Islands (Zone 1, Figure 1). This is directly north of the
island of Hispaniola. Although separate geological formations, the
Turks and Caicos and the Bahamas are very similar physiographically.
In terms of geological composition, terrain and vegetation the Turks
and Caicos are comparable to the Bahamian islands of the Inaguas and
Mayaguana. The drier islands, i.e., Grand Turk, Salt Cay, South
Caicos, East Caicos and West Caicos resemble the former. Middle
Caicos, North Caicos and Providenciales and associated cays are
similar to Mayaguana in that they have sufficient rainfall to support
a moderate stand of coppice vegetation. All the islands are sedi-
mentary formations and are therefore geologically distinct from the
neighboring Greater Antillian islands of Cuba and Hispaniola which
are largely igneous and metamorphic formations.

S .. C. ........!

West Caicos
--... ..

..French Cay /!South Caicos -'
S .i: Turk 1

"oWest Sand Spit o .
.'... An r Sal.Cay
.**.. .* Bush Cay
9 Sand Cay

Smiles --

Figure 2. Turks & Caicos Islands


Swidden agriculture is practiced in the Turks and Caicos only
on the wet group of islands. Subsistence agriculture is impossible
on the dry islands, except during periods of unusually heavy rainfall
which occur with an approximate average of once in every 10 years.
Conversely, drought jeopardizes crops on wet islands at least once a

Research Overview

It should be noted that at the time of this writing approximately
70% of the artifacts recovered during the field research have been
formally analyzed. Even though final percentages are not available
on many of the distributions, enough analysis has been completed for
distinct patterns to emerge. It is anticipated that the final figures
will not indicate patterns significantly different from those described

The combined number of sites located during DeBooy's 1912 survey
and those of the 1976-78 surveys are shown in Table I. The table
includes only those sites that he actually visited. Sites which were
reported to him by local residents but not visited are not included
since our experience with local guides in these islands shows that
local residents do not commonly differentiate between historic glazed
ceramics and aboriginal pottery. Their reports of the location of
camps of the "Olden Time People" or of "Indians" must be evaluated by
on site inspection. There is usually a site where they report one,
but it is virtually always a historic European occupation.


ISLAND DeBooy's 1912 Survey 1976-78 Surveys Total:

Middle Caicos 4 35 37*
North Caicos 4 0 4
Providenciales 3 4 7
Pine Cay 0 1 1

Grand Turk not surveyed 0 0
South Caicos not surveyed 0 0
East Caicos 2 0 2
Iguan Cay (near not surveyed 1 1
E. Caicos)

This total assumes two duplications between DeBooy's
survey and the 1976-78 surveys.



As the preceding table illustrates, the prehistoric occupa-
tion of the Turks and Caicos Islands was concentrated on the island
of Middle Caicos. Consequently this was the focus of our research.

Thirty-five aboriginal sites were discovered on Middle Caicos
during our field studies. Two of these may be identical to two of
the sites reported by DeBooy (1912). The site he described on the
"salina" southwest of Lorimers (Ibid: 101) is possibly the same as
a site we designated MC-6. One glaring contradiction to this pos-
sibility is the fact that DeBooy reported finding a total of 17 sherds
at the site he visited, while we collected over 8,000 sherds from
the surface of 1,200 sq. m. in the most accessible portion of the
site. It is equally difficult for us to accept that DeBooy visited
MC-6 and could find only 17 sherds and that in our surveys we missed
a substantial site with mounds as described by DeBooy.



~~'~ 47 ~:S:.~.~:r
,:' : P ~
r I .+..:






- --- -; L tIDAL PLATS






Figure 3. Middle Caicos

*" '* **


The site DeBooy located on Dead Man's Skull Bluff near the
village of Conch Bar (1912: 100) is probably identical to one of
two sites, MC-1 or MC-2, which we surface collected during our sur-
vey. We never located the sites reported by DeBooy (Ibid: 102) on
Gamble Hill and Indian Hill, north of the settlement of Lorimers, so
the current total of sites reported for Middle Caicos is at least 37.

It is rather striking that of the fifty-two sites recorded for
the Turks and Caicos Islands thirty-seven are on Middle Caicos. Dur-
ing our survey only seven aboriginal sites were found on neighboring
islands. Middle Caicos has slightly less than one third of the total
land area of the Turks and Caicos. One cannot assume a random dis-
tribution of sites when over 70% of them occur on less than 33% of
the available land. Indeed, prehistoric occupation concentrated on
the eastern half of Middle Caicos where twenty sites are situated.
These sites were the locus of such dense and long term occupation
that over 95% of the artifacts obtained from the Caicos in our surface
and subsurface collections, and those of DeBooy, came from this group
of sites. Site Mc-6 accounts for more than 90% of that total.

Sites With Only Imported Ceramics

In the Bahamas no sites have been reported to date that have
a predominate amount of igneous based ceramics imported from the
Greater Antilles. All the known Lucayan Arawak sites have ex-
clusively or predomantly locally made ceramics employing native
Bahamian sedimentary based clays and shell tempering which has
been referred to as Palmetto Ware. If it is ethnographically
correct to refer to the Bahamas and Turks and Caicos as the
"Lucayan Islands", then by extension they are, archeologically
speaking, the "Palmetto Ware Islands." Every island in the group
on which Arawak sites have been documented has Palmetto Ware

The same cannot be said for every individual site in the Caicos
Islands, however. Of the 41 sites recorded and analyzed by the
writer, there are twenty-six sites that have only imported
ceramics. Twenty-five of these are on Middle Caicos and one is on
Providenciales. These are small sites that are less than hectare
in area and are camps or small "hamlets", in the sense of the phrase
as employed by Blanton (1972). These are represented in Figure 3 by
small unnumbered black dots. Stylistically, the ceramics from these
sites are early-middle Meillacoid, probably of the Meillac style
(Rouse 1941: 54-91) from northwestern Hispaniola. These sites would
therefore most probably date between 750 and 900 A.D. Table 2
indicates the environmental zones that are associated with these
sites. They were found in the following environmental contexts:



Environmental Zone: Number of Sites: Island:
Inland hilltop/hardwoods 5 Middle Caicos
Coastal hilltop/scrub 2 Middle Caicos
North coastal plain/scrub 4 Middle Caicos
Savannah 1 Middle Caicos
Savannah hardwood ecotone 2 Middle Caicos
South coastal tidal flat/hardwood ecotone 0 Middle Caicos
Margin of salt water inlet 3 Middle Caicos &

Initially, this concentration of imported ceramic sites looks
precisely like what one would expect the process of Arawak "discovery
and exploration" to look like archeologically. Obviously, the Antil-
lian Arawaks at some period of time had to discover and explore these
islands and leave material evidence. We know that as a general trend
the direction of exploration and colonization of the Antillian Arawaks
was roughly from east to west in the northern Caribbean. The fact that
comparable "discovery and exploration" sites are unknown from the
northwestern portions of the Lucayan Islands, i.e., Zones 2 & 3 shown
in Map I, inclines one toward the hypothesis that these islands in
Zones 2 & 3 were the recipients of colonists from Zone I where the
processes of discovery and adaptation to the relatively impoverished
environment of the Lucayan Islands had already occurred. Indeed, that
is an assumption that William Sears and this author stated in an earlier
writing (1978: 22-3).

There are problems with this conceptualization however. As will
be discussed below, there is a more complex explanation for the high
frequency of imported ceramics in the Caicos Islands, especially on
Middle Caicos. Additionally, the manufacture of Palmetto Ware does
not, strictly speaking, document shared cultural heritage among the
manufacturers. There are very real physical constraints upon ceramic
manufacture in the Lucayan Islands that impel a certain degree of
homogeneity in the results. With few exceptions, for example, Bahama
Red Loam is the only local soil with the requisite plastic qualities
for ceramic construction and that only in combination with the struc
tural strength imparted by laminar shell tempering. This crude mat-
erial is not well suited to modeling which restricts artistic license
in this medium. Additionally, our experiments indicate that the thick
vessel walls required for structural integrity with this paste and
tempering are not often self supporting when being molded and dried
prior to firing. One compensatory construction strategy is to mass
mold the lower half of the vessel in a basket which is not removed
until after the paste dries. This would cause the basket weave im-
pressions that are so distinctive of Palmetto Ware. It also provides
an explanation for the impressions on curved surfaces which had moti-
vated previous speculation (Sears and Sullivan 1978: 13).


From the above, some might hypothesize that if there was ongoing
colonization of the Lucayan Islands by Antillian Arawaks from, most
likely, Hispaniola and Cuba, then shortly after their arrival the
bearers of disparate Arawakan ceramic styles would be producing Pal-
metto Ware throughout the archipelago because of the material con-
straints rather than due to common cultural heritage. Unfortunately,
we are not presently in a position to answer this question of the
relative influence of material coersion and current cultural style
because we lack comparative statistical data on such stylistic
variables as shoulder, neck and rim forms, and the decorative modes
of punctation, incision and simple modeling, for Lucayan sites in
the Bahamas. Similar data will be presented in future papers on our
samples from the Caicos Islands. The labor intensive program of re-
examining the Bahamian collections on this basis should probably await
the discovery of a single Bahamian site where imports predominate or
where Palmetto Ware are altogether lacking. Until that time, the
prime candidate for the region of the initial Arawak interaction with
the Lucayan islands lies in Zone I in the Caicos Islands.

It was stated earlier that there is a more complex explanation
for the presence of sites with only imported igneous based ceramics
than just the detritus of discovery and exploration. The broader
interpretation is based on the physical distribution of these and
related contemporary and more recent sites relative to the island's
resources and the implied form of exploitation of the islands by
Antillian Arawaks. A simple theory of outward colonization from
Hispaniola in response to population pressure and competition for
scarce resources might apply neatly to the contemporary expansion into
Cuba, but is not entirely satisfactory to explain the early archeo-
logical record in the Caicos Islands where colonization appears to
long postdate exploitation. The more complex explanation can only be
properly approached by considering another natural grouping of sites
which will now be addressed.

Sites With Predominantly Imported Ceramics

Three sites, MC-8, MC-10, and MC-33 (Figure 3) have imported cer-
amics primarily and also have Palmetto Ware. About 70 separate sur-
face collections were made at these sites. Palmetto Ware sherds were
recovered with a frequency ranging from 0-12%.

Each of these sites is over 100 meters in size, large by regional
standards. In terms of area, MC-8 and MC-10 are approximately one
hectare each. The original size of MC-33 is difficult to determine
because the site has been heavily disturbed by storm wave action. Of
these sites, all have produced substantial ceramic samples with the
MC-8 collections totaling just over 900 specimens, MC-10 just under
1,500, and MC-33 nearly 200. Examples of ceramics typical of these
sites are shown in Figures 4 and 5. Although no stratigraphic data
were obtained from these sites, the ceramic stylistic variation, trac-
able as a progression, indicates that they are multicomponent. Ele-
ments of this progression are discussed below.



MC-8 and MC-10 are only 70 meters apart. They are situated in
patches of guinea grass and cabbage palms at the juncture of the hard-
wood forest and tidal flats along the southern edge of the island.
Nine of the small sites with only imported ceramics occur in this
environmental association.

Following the lead of Vita-Finzi and Higgs (1970), it is appro-
priate to consider the resources exploitable by the Arawaks in rela-
tion to associated "catchment areas" as possible motivation factors
for site selection. Nine sites included among the two site categories
discussed thus far, including MC-8 and MC-10, are situated like a
corona around the most conspicuous natural resource on Middle Caicos,
Armstrong Pond. Armstrong Pond is a natural salt pan that produces
hundreds of tons of raw crystalline salt annually. It is the largest
of five natural salt pans in the eastern half of Middle Caicos. The
nine sites mentioned are all within a twenty minute walk of Armstrong
Pond. MC-33 and five of the remaining exclusive ceramic sites have
one or more of these salt pans within a similar radius or "site catch-
ment". Before considering the subject of salt in more detail the
artifact content of these sites will be examined.

The import dominant sites are characterized by ceramics with fine
line incision and punctation similar to that present at the import
exclusive sites (e.g., Figure 4, number 208, upper left; Figure 5),
as well as incised applique strips and nubbins, loop handles, strap
handles, red slipping, modeled lugs, and a few examples of broad line
incision and curvilinear incision (See Figures 4& 5). These last two
modes indicate some Chicoid (Rouse 1941: 113-21) stylistic influence.
These specimens appear to be transitional between the Meillacoid
ceramics found at the import exclusive sites and in most of the col-
lections from MC-8, MC-10, and MC-33, and fully developed Chicoid
ceramics found among imports at sites to be described later.

On the basis of stylistic analysis of the many discreet collections
from import dominant sites on Middle Caicos, the following tentative
reconstruction of the ceramic stylistic progression was derived for
the period prior to the occurrence of Palmetto Ware dominant sites.

Phase I: Chronology: 750-800 A.D.(?)

Forms: Simple hemispherical, globular and subglobular
bowls with direct or incurving shoulders, teco-
mated, platters and griddles.
Rims: Simple and direct, normally rounded or flat, not
Decorations: Usually none, but many specimens have punctation
just below the rim on the exterior. Simple effigy
lugs occur. They might be termed surreal or
deimorphic because they probably represent the super-
natural spirits of the Antillian Arawakan cosmology,
known generically as zemis.




Palmetto Ware:

Phase I examples:

The dominant temper type is a very distinctive
dark ferro-magnesium mineral or group of related
minerals, usually 1 to 4 mm in diameter. Large
amounts of it were used so these ceramics are
very distinctive. Less common is angular igneous
grit which includes a wide variety of mineral type,
fragment sizes and temper to paste ratios.

Either very little or no Palmetto Ware was made
during this period. The Palmetto Ware found on the
surface of these sites does not include the decor-
ative modes of Phase I, but rather those of Phase II.

Collection units 208 (except 2nd from right top).
210 and 215, above; 190, 191, 173-2, Figure 4.

Figure 4. Phase I Ceramics. Site MC-10.

hnea TT.

Chronoloov 800-900 A.D.(?)

Forms: Same as Phase I plus carinated and concave shouldered
vessels plus "boat shaped" vessels and bottles.
Rims: Direct, flanged or folded in or out, outflaring,

Nw- w .----

K AC&*-'AI-- L 9%N &am%0& T oma %F n - -- 1





Punctation, fine line incision, incised and punc-
tated applique strips or nubbins, modeled lugs
with doughnut eyes (as in Phase I), loop handles,
strap handles red slipping.
Same as Phase I plus quartz sand tempering and
virtually temperless ceramics. The latter are
finely finished.

Palmetto Ware:
Forms: Ware was simple and probably experimental during
this phase.
Rims: Much simpler than the imported ceramics. Direct
or incurving, not folded or flanged. Lips rounded
or flattened.
Decorations: Same as Phase II imports but simpler designs and
very little modeling.
Tempering: Crushed clam shells (Codakia orbicularis)
Phase II examples: Collection units 192, 182, 77, 78, 193, 180, 4,
183, 190 (left), above; 212, 213 and 209 (top 2nd
from left). Figure 5.



^ ^ 'f ^ ...-'. ..... ... ... .... .. .. .. .... .. ... .
OK -12-

Figure 5. Phase II Ceramics. Sites MC-4
(#4) and MC-8



Phase III: Chronology 900-925 A.D.(?)
Forms: Same as Phase II but less carination, more simple
Decorations: Same as Phase II with the addition of broad line
curivilinear incision and cessation of modeled lugs
with doughnut eyes.
Tempering: Decrease in ferromaganese tempering, increase in
sparse igneous grit tempering and in virtually
temperless ceramics.
Rims: Less frequently flanged or folded. Usually direct
or incurving

Palmetto Ware:
As this is a very short period and the sample is
very small, no realistic judgment can be given, but
probably similar to Phase II.

Phase III examples: Figure 4; 212 (lower left), 209; Figure 5; 187.

It is possible to conclude that Palmetto Ware first appears in
the Caicos Islands contemporary with a derivative from Meillacoid
foundations because all the ceramics of Phases I and II are unequi-
vically Meillacoid.

Ostionoid ceramics, or something very similar, appear on Middle
Caicos after Palmetto Ware production had begun. Initially this
appears illogical since in any given area of Hispaniola Ostionoid
ceramics are likely to preceded rather than follow Meillacoid. The
explanation for this is hypothesized to be the result of the expansion
of the trade network to include new areas where a different timetable
for the stylistic progression prevailed. Although some Ostinoid in-
fluence on early Palmetto Ware makers is possible, it does not appear
to have had as strong an influence as proposed in an earlier writing
previous to this author's investigations at MC-8 and MC-33 (Sears and
Sullivan 1978: 23).


The dominance of imported utilitarian artifacts which have a
relatively high consumption (breakage) rate, i.e., ceramics, in the
sites described above, suggests seasonal visitation. The nearest
possible source of these ceramics is over 100 miles away and must be
acquired through a labor intensive canoe trip. The pressure for the
development of an indigenous ceramic production would have been much
greater than the archeology of the Middle Caicos suggests if seasonal
visitation was not the practice the logical source of visits is
northwestern Hispaniola, specifically, the aboriginal province of
Marien (Rouse 1948: 529). This conclusion is based on both nautical
data and stylistic similarities. The same probable derivation can be
assigned for the trade ceramics we located on Pine Cay and on Provi-



Seasonal activities are usually geared to the exploitation of
resources available during a specific period of the year. Weather
permitting, in the Turks and Caicos Islands, fish are available but
they are also available in the Antilles. Domestic and wild plant
foods in these islands are seasonal but better opportunities for
exploiting agricultural and plant food resources are available in
the better watered and deeper soils of the Antilles. Conchs occur
in greater quantities in the Turks and Caicos Islands than in the
Antilles, and they are generally available throughout the year, the
only exception being when they bury themselves during winter cold
spells and storms. There is only one resource available on a seasonal
basis in the Turks and Caicos Islands which appears likely to have
motivated seasonal visitations from the Greater Antilles. That resource
is salt. Historically, salt occurred in the summer dry season in nat-
ural salt pans on all the major islands of the Turks and Caicos.


Although Middle Caicos is not the only salt producing island in
the group, it is the closest island to the Greater Antilles that pro-
duced large amounts of salt and also had a reliable fresh water supply.
Fresh water and relative proximity to the ports of embarkation seem to
be the key to the choice of Middle Caicos over other available salt pro-
ducing islands. On Middle Caicos, the distribution of salt pans appears
to be the major cause for the concentration of settlements in the east-
ern half of the island.

The largest salt pan on Middle Caicos is Armstrong Pond. This
lake is nearly a mile long and is situated among hardwoods 600 to 1000
meters from the edge of the southern tidal flats. In the late summer,
strong trade winds and a diminished rainfall conspire to evaporate the
majority of the water, leaving tens of thousands of bushels of crystal-
line salt. The pond is a renewable resource since it is constantly
replenished by salt springs energized by the tides.

As an experiment, 16 people gathered salt in Armstrong Pond in
late July, 1977, for 15 minutes. Conch shells were used as scoops.
In that period, 120 gallon containers of salt were gathered. If we
assume a sustainable rate of half that demonstrated during the experi-
ment and use a six hour day for the sake of argument, then that crew
could have gathered over 1300 gallon containers, or in dry measure,
over 139 bushels of salt.

A normal western diet provides an average of 9gofNaCl per day
which is adequate for metabolic needs for those not physically active
in temperate climates; on the other hand, doing heavy work in the
tropics may require as much as 30 g of NaCl to make up body loss
(Davidson, et. al. 1975: 101, 664). Arawaks probably ate less natural
salt from every day meals such as mammal meat, than did modern western-
ers. They certainly lived in a hot climate in the Antilles and pre-
sumably were frequently engaged in heavy labor. Logically, they had
a metabolic need for a salt supplement in their diet independent of


any taste preferences which might have motivated such an addition.

Theoretically, the Arawaks could have evaporated salt from sea-
water on any coast. It was probably much less labor expensive to go
where nature had already done that work. Moreover, it would have taken
a very long time for the Antillian Arawaks to produce 139 bushels of
salt through evaporation of saltwater in small containers. It was
probably prohibitively labor expensive to produce enough salt in this
manner to satisfy the needs of 100,000 to 600,000 Arawaks on Hispaniola
(Rouse 1948: 522), the presumed source of seasonal visitors.

The hypothesis that the Arawaks considered salt a valuable resource
on Middle Caicos is supported by the circumstantial evidence of the
clustering of sites in eastern Middle Caicos in the vicinity of the
salt pans and in reference to no other obvious motivation. Additional
evidence is provided by the location of a major site, MC-6, in that
same area.

Lucayan Sites

The social, economic, and political intricacies of the shift from
seasonal exploitation of the resources to actual colonization of the
Caicos Islands, although intriguing, will not be discussed here. How-
ever, at some point contemporary with late Meillacoid styles in north-
western Hispaniola the processes began on Middle Caicos. This develop-
ment is reflected archeologically by the flowering of a truly indige-
nous ceramic assemblage, Palmetto Ware, and by an expansion in the
trading network shown by new varieties of imported ceramics, as well as
by new types of settlements.

On Middle Caicos all but one of the Lucayan sites are on or near
the north coastal plain. Here were found two large sites, ca. one
hectare, which follow the classic pattern of Lucayan villages in the
Bahamas. They are near a sandy beach on a protected shore in the vicin-
ity of rich shellfish resources (Sites MC-12 & MC-32, Fig. 3).
Lucayan sites following this general pattern were also found on Pine
Cay (between North Caicos and Providenciales) and on Providenciales.
Test excavations at the two Middle Caicos sites indicate that they are
long term permanent settlements. Throughout their occupation they had
access to simply decorated Meillacoid and Chicoid imports from the

Two small sites, less than 1/4 hectare, also show a significant
proportion of Palmetto Ware, but the samples are too small to be
statistically valid. One is near shore on the north coastal plain
and the other is on a hardwood covered hilltop overlooking the coastal

Two Lucayan sites were found in Conch Bar Cave. One was small
and disturbed and surface collected only. The other, miraculously,
appeared completely undisturbed. Test excavations provided a fan-
tastically rich faunal sample, the preservation in the cave being



superb. The cave sites showed the same relative percentage of imports
as the large sites on the north coast.

Site MC-6

Site MC-6, the most fascinating of the Lucayan sites, is in
guinea grass and palms at the juncture of the hardwoods and tidal
flats (Figure 3), an eco-niche identical to that of MC-8 and MC-10.
MC-6 is the most complex site reported to date from the Lucayan
Islands. It covers an area of approximately 1.8 hectares.

The dominant ceramic tradition is the locally made Palmetto Ware
which constitutes approximately 92% of the collections. The remainder
of the ceramics are imports from a variety of traditions, some examples
of which appear in Figures 6 and 7. The trade ceramics include late
Meillacoid, full Chicoid, and intermediate styles.

Figure 6. Ceramic Sherds from MC-6
(Plaza 1)

There are two additional styles that are difficult to identify
with assurance. One appears to be a late surviving Ostionoid style
with simple bowl forms lacking any decoration with the exception of
red slipping. The other is a fine white to gray pasted ware with
very fine finish which could easily be classified as historic if it



were not for the stratigraphic context it were associated with, and the
fact that one example has a double row of punctations, an aboriginal
design motif, on the interior of the vessel.

Figure 7. Ceramic Artifacts from MC-6
(Plaza 1)

Whereas the imported ceramics of the import dominant and import
exclusive sites appear to be from northwestern Hispaniola reflecting
a period of about 925 A.D., the trade ceramics at MC-6 appear to come
from traditions widely dispersed among the Greater Antilles and perhaps
the Virgin Islands. They probably range in date from around 925 A.D.
to the time of Spanish contact. It seems to be an inescapable con-
clusion that MC-6 was a component, if not the focus, of a very
widespread trade network.

The imported ceramics at MC-6 could be the subject of a dis-
sertation in themselves. Surface collections alone from MC-6 contain
over 30,000 sherds. In the present discussion these ceramics will be
dealt with in a limited manner. Only the distribution of exotic types
among residential zones of the community will be considered.

MC-6 is a planned community. It consists of two residential groups
surrounding plazas joined end to end (Fig. 8). Plaza I is approxi-
tely 160 meters long; Plaza II is just under 100 meters in length. The



plazas are distinguished by the absence of most loose surface rock and
by the relative scarcity of occupational debris. The plazas are sur-
rounded by ridges and mounds of midden mixed with stone which appar-
ently was used in the construction of the walls of structures.

The western plaza, Plaza I, has within it an artificially leveled
area flanked by rows of undressed stone and low ridges. The dimen-
sions of this feature are approximately 16 by 30 meters. Although
other concurrent functions are probable, this feature is identical
in format to Arawakan constructions in Hispaniola and Puerto Rico
which are considered to be ball courts. It appears therefore that
the statement that the Lucayans lacked ball courts made by DeHostos
(1948: 544) is, in at least one instance, incorrect. This possible
ball court is but one of several exceptional features at MC-6.

Figure 8. Site MC-6



Young (1977: 384) notes that, "Rome's major highway was the Via
Salaria (Salt Road). Over it soldiers conveyed the precious crystals
up the Tiber from the salt pans at Ostia." Middle Caicos appears to
have had its own "Salt Road." Extending approximately 1/3 mile from
the north margin of Plaza I to the edge of the maximal floodplain of
Armstrong Pond is an aboriginal road. The surface of the road has
been cleared of most surface stone and the margins of the road are
flanked by a stone ridge 30-120 cm. high and 40-200 cm. wide. The
road averages 6 meters wide.

Plaza I is surrounded by a discontinuous midden ridge up to one
meter high surmounted by eight definite and fourteen probable remains
of semi-pit houses that employed undressed limestone in the founda-
tion and walls. These house features are probably identical with the
"mounds" described by DeBooy (1912: 97) which the writer found dif-
ficult to credit as aboriginal features before seeing them (Sullivan
1974: 8). One of these structures, which is situated at the juncture
of the two plazas has two chambers and two entrances. One of these
entrances and chambers is aligned with a portion of the northern bound-
ary line of the ball court. The possibility of an astronomical deriva-
tion for this and other alignments at the site is being investigated.

Plaza II contains no ball court and is less well cleared of loose
rock. This plaza is also flanked by a midden ridge that has along it
15 mounds believed to be structural remnants.

Residential Zones at MC-6

Based on differential access to imported stone and certain types
of ceramics by the prehistoric population it appears probably that
there were three distinct residential zones at MC-6. Each hypotheti-
cally corresponds to some social grouping in which membership was
probably based upon kinship ties. The differential access is based
on recovered data in the archeological record. The social/kinship
groupings are, of course, interpretation.

These differential artifactual assemblages manifest themselves
as follows: Elaborately decorated imported Chicoid and late Meil-
lacoid ceramics are restricted to the midden and structures around
Plaza I. Contemporary but less elaborately decorated Chicoid and late
Meillacoid ceramics are found among the midden and structural debris
around Plaza II. Deimorphic or "zemi" lugs, such as appear in Figure
7, are restricted to the Plaza I area, as are the two unidentified
imported ceramic types described earlier. Over 90% of the imported
igneous stone found at the site, rough or worked, comes from the
Plaza I grouping. White slipped ceramics, presumably Chicoid, are
found only in Plaza I. The remains of semi-pit houses are found only
around Plaza I. Decorated and worked shell has been found only around
Plaza I. Obvious public works, roads, etc. are found in Plaza I.


Thus far the evidence suggests only two distinct residential
groups, those of Plaza I and those of Plaza II. Indeed, that is the
primary distinction, but the distribution of artifacts permits a
more refined segregation. Within the Plaza I grouping, the inhab-
itants of the southern flank of the Plaza appear to have been of
higher status. In this southern zone were found most of the zemi
lugs, all cut olive shell (Oliva sayana) beads, the only examples of
stone celts and decorated shell, the only chert tool, most of the
white slipped imported ceramics and virtually all of the unidentified
rare ceramic imports. Most of the unmodified igneous rough rock
comes from this area also.

Some of the concentration of material in the southern zone can
be explained by proposing that it was the residential area of craft
specialists. Indeed, this appears to be the case with reference to
the shell beads and decorated shell. The raw materials are locally
available however, so access to imports is not necessarily a relevant
consideration. The lone chert tool found in this area, possibly an
import, appears to be suited for making the characteristic cut in the
olive shell beads which permits them to be strung or suspended. It
might also be argued that the concentration of igneous rough rock
and the occurrence of the only celt fragments in this area reflects
ground stone tool manufacture. This is an open question as far as
the writer is concerned. There is no obvious debris from such a hypo-
thetical activity, just possible raw materials and a few examples of
finished tools. There are no chip concentrations or partially completed
tools as one might reasonable expect from such an activity area. It
seems quite possible that hand sized igneous cobbles, probably from
stream beds, were being imported for direct use without modification.
Igneous sandstones also were brought in in much the same manner.

If the artifact distributions at MC-6 are accepted as status
indicators, and status groupings as ranked kin based residential
groups, then Plaza I and Plaza II can be modeled as upper and lower
moieties. Carrying the analysis further, the northern and southern
residential groupings of Plaza I can be interpreted as ranked line-
ages; the southern group being of higher status.

It is not without coincidence that this model sounds like the
Taino Arawak distinction between nobles and commoners (Rouse 1948;
529-530). This system might have included a recognized chief's
lineage among the nobles.

The artifacts and faunal material recovered from the centrally
placed two chambered structure at MC-6 indicate that it was a house,
However, it was an exceptional house owing to its location, size and
design. Among the Taino the chiefs were entitled to large "special"
houses which were frequently situated at the end of a ball court
(Rouse 1948: 525, 529). The two chambered structure at the juncture
of the plazas appears to be a worthy candidate for a chief's house.
The "chief's house" is physically continuous with the highest status
residential group, the south side of Plaza I. It is possible that
this residential group owes its relative status to a kinship connect-


ion with the chief, and is the residential locus of the chief's
lineage. This might be offered as an explanation for preferred
access to scarce imported materials which were presumably highly

Arawak chiefs were believed to have supernatural allies, or
zemis, which were the most powerful zemis of the village, and
he (the chief) organized their workshop by the villagers."; to that
end some ". chiefs had additional buildings outside the village..
(including) a temple to house his idols." (Rouse 1948: 525).

There is only the one structure associated with MC-6 which has
two chambers like the central structure hypothesized to be the chief's
house. It adjoins the west side of the road approximately halfway
to Armstrong Pond. Surface collections and test excavations indicated
that this structure was different than those surrounding the plazas.
There were no faunal remains of any sort found, and only one imported
aboriginal sherd. The structure was, evidently, functionally distinct
from those at MC-6 proper. As noted, it is formally related to the
"chief's house." The idol houses of the Taino are believed to have
been separate from the village, identified with the chief, and func-
tionally distinct from other structures. The two chambered structure
beside the road could be many things, but is certainly a reasonable
candidate for an idol house associated with the zemi of the chief of

The road which runs alongside the structure in question leads
from the high status plaza to a vast source of salt. As with many
other sites previously mentioned, salt is probably the prime mover
in determining the location of MC-6. The exportation of salt to the
Greater Antilles could be the resource stimulating the import of a
wide array of items at MC-6 not found at any other Lucayan site in the
region. Salt and derivitive trade are the logical source of an econ-
omic surplus at MC-6. Based upon what we know of the Arawak cosmology
and social structure, it is difficult to imagine a chief of MC-6 whose
personal zemi was not conceptually bound to the source of the village's
prosperity, i.e., salt. A circumstantial case can therefore be made
for the function of the roadside structure being a shrine to the zemi
of the chief, the salt zemi.

Although the early seasonal visitors to Middle Caicos were pre-
sumably the emissaries of chiefs and chiefdoms on Hispaniola, there
is no obvious social or economic stratification in evidence within or
between these sites. Stratification is demonstratable only after col-
onization has occurred and the new indigenous culture, the Lucayans,
are in apparent control of the salt pans which are the wellspring of
an economic surplus. The distinctly Taino look of MC-6 is in contrast
to the commonly assumed "Sub-Taino", i.e., simple chiefdom, level of
the Lucayans generally. This is probably explainable in the context
of frequent trading contact with the Taino and through intermarriage
with the Taino to solidify commercial ties. Assuming the genetic and
cultural flow followed the tendrils of the trade network, the Lucayans
of MC-6 would naturally adopt the social structural forms of the Taino.



This is a more plausible explanation than any sort of scheme of
convergent evolution.

There are, however, probably some evolutionary nuances involved
in the developments on Middle Caicos. Imitation can explain only
some of the particulars of similarities between a Lucayan chiefdom
seated at MC-6 and the Taino. There is a structural substratum which
must exist before cultural transfers can be made,

In seeking an explanatory framework for the social stratification
and complexity of MC-6 many of the concepts subsumed under the heading
"Gateway Community" as delineated by Kenneth Hirth (1978: 35-37) are
appealing. He states in part:

"The emergence of stratified societies with perpetual leader-
ship appears to have been closely related to control over the
production and redistribution of resources."

"An increase in long distance trade occurs with the
appearance of incipient chiefdom societies, thus sug-
gesting that trade is an important factor in the process
of social stratification. The raw materials used in
enhancing and reinforcing the statuses of ranked soci-
eties were obtained through trade."

"Gateway communities develop either as a response to
increased trade or to the settling of sparsely populated
frontier areas. They generally are located along natural
corridors of communication at the critical passages between
areas of high mineral, agricultural or craft productivity.
They often occur along economic shear lines where cost
factors change and there are economic discontinuities in
the free movement of merchandise." (emphasis added)

This author believes that there are many congruences between
Eirth's concepts and MC-6. For example, the leadership, or elite,
of MC-6 would have had control over the gathering and exportation of
salt. Distinct social stratification appears coeval with evidence
for an expanded long distance trading system which first appears on
Middle Caicos with the settlement of MC-6. Raw materials, e.g., stone,
apparently obtained through trade, flowed to particular residential
zones and social groups and were apparently social status markers.
The site appears to have been the focal point for settling of a
sparsely populated frontier area. MC-6 is located between natural
boat landing sites for traders arriving from Hispaniola and a zone of
high mineral concentration. There are many more similarities between
the concept of a Gateway Community and the traits of MC-6, but those
will receive more lengthy treatment in future publications.

I would like to consider the archeological data discussed in
relation to a more general scheme of the Arawak expansion into the
Caicos Islands, There may be some more generalized applications for
this scheme relative to Arawak expansion in the Caribbean as a whole
as well.



Some consideration should be given to the impetus for arrival of
the Arawak in the Caicos Islands. There must have been a period before
actual contact when the presence of the islands was known or suspected.
Knowledge of these islands might have resulted from reports by non-
agricultural peoples of Hispaniola, the Ciboney, who were absorbed or
displaced by the Arawaks. Other indicators would have been wave re-
flections from the Caicos Bank detectable off the north coast of
Hispaniola, bird migration patterns, especially flamingoes, and pro-
bably the most reliable; repeated lightning storms at night occurring
over the islands. These storms are caused by convection cells over
the sun heated landmass and are visible for well over a hundred miles.
Hispaniola and Cuba are "visible" in this manner from the Caicos
Islands. So it is reasonable to hypothesize that the first Arawak
interaction from Hispaniola with the Caicos Islands resulted from
observations of natural phenomena indicating that there were more
islands to the north.

Actual contact or "discovery" might have been accidental, as in
the case of voyagers blown off course by storms, but more likely, it
was intentional. Motivation for intentional voyages of discovery in
the Caribbean as elsewhere could and probably did run the gamut from
curiosity and adventure-seeking to a quest for new lands to vent
increasing population, social pressures, or for economic exploitation.

Once discovered, these islands would have been subject to a period
of exploration and assessment of the new territory. This phase pro-
bably produced the first archeological evidence. Some of the import
exclusive sites mentioned would be associated with this time period.

When the first Arawak explorers returned to their villages their
assessment of the Caicos Islands must have been less than glowing.
Compared to Hispaniola and the other Greater Antilles they were un-
attractive for settlement. The Arawak explorers would have returned
with negative reports on agricultural soil quality, fresh water avail-
ability on most islands, clay sources, trees for canoes, stone for
tool manufacture, andperhaps, not unimportantly, mosquito and sand
fly populations. These islands would have been probably very low
priority areas for actual settlement. However, the islands did have
salt, abundant fish and shellfish, adequate fresh water in selected
cases, and significantly, no inhabitants. The salt could be exploited
without local human interference or competition.

So while Cuba and Jamaica were being colonized the Caicos were
merely exploited seasonally during periods when the salt was available.
This time period is reflected archeologically as some of the import
exclusive sites and the import dominant sites, MC-8, MC-10 and MC-33.

Colonization did eventually occur, but it is doubtful that it was
in response to exaggerated population or social pressures for outlets
from Hispaniola. Rather, it is quite possible that colonization was
first initiated in order to maintain claim to the salt deposits and



to prevent outside interlopers from threatening the monopoly of the
original discoverers. The colonists eventually expanded their trad-
ing network to areas outside northwestern Hispaniola. The material
from MC-6 seems to testify to this development.

As a result of the combined influences of adaptation to the
new environment and a degree of isolation from the parental culture,
a new culture developed; that of the Lucayans. As the population
expanded and control of the salt trade became concentrated in the
hands of a local elite, subpopulations fissioned off and began set-
tling other portions of the Caicos Islands and Bahamas. The local
economic adaptation as well as the difficulties of integrating with
established populations would have disposed them toward settlement
in this new direction rather than back into the Greater Antilles.
However, as was suggested earlier, the settlement of the Bahamas
might well have been augmented by infusions of people from Cuba.


It seems worthwhile to regard Arawak expansion in the Caribbean
as the product of selective and tailored exploitation of the resources
of individual islands and island groups rather than simply the result
of overflow from previously held territory. The resources of a new
island, such as Middle Caicos, may have been integrated into the
economic system of a population long established elsewhere. Explora-
tion may have served the purpose of providing the sponsoring peoples
with a wide spectrum of resource exploitation options.

References Cited

Blanton, R.E.
1972 Prehispanic settlement patterns of the Ixtapalapa
peninsula region, Mexico. Occasional Papers in
Anthropology No. 6. Dept. of Anthropology, Pennsyl-
vania State University, University Park.

Davidson, Sr. S., P. Passmore, J.F. Brock, A.S. Truswel
1975 Human Nutrition and Dietetics. Sixth edition.
Churchill & Livingston. London.

DeBooy, Theodor
1912 Lucayan remains in the Caicos Islands. American
Anthropologist, 14(1): 81-105.

DeHostos. A.
1948 The Ethnography of Puerto Rico. Handbook of South Amer-
ican Indians, edited by J.H. Steward, Bureau of American
Ethnology, Bulletin 143, Vol. 4, Pt. 3, pp. 540-546.

Hirth, K.G.
1978 Interregional Trade and the Formation of Prehistoric
Gateway Communities. American Antiquity, 43(1): 35-45.



Rouse, I.
1941 Culture in the Ft. Liberte region, Haiti. Yale University
Publications in Anthropology. No. 23. New Haven.

1948 The Arawak. In Handbook of South American Indians. Edited
by J.H. Steward, Bureau of American Ethnology, Bulletin 143,
Vol. 4, Pt. 3, pp. 507-546. Washington.

1977 Pattern and process in West Indian archeology. World
Archaeology, 9(1): 1-11.

Sears, W.H., and S. Sullivan
1978 Bahamas Prehistory. American Antiquity, 43(1): 3-25.

Sullivan, S.
1974 Archaeological Reconnaissance of Eleuthera, Bahamas.
Masters Thesis. Florida Atlantic University, Boca Raton.

Vita-Finzi, C., and E.S. Higgs
1970 Prehistoric economy in the Mt. Carmel area of Palestine:
Site catchment analysis. Proceedings of the Prehistoric
Society, 36: 1-37.

Young, G.
1977 Salt: The Essence of Life. National Geographic,
152(3): 381-401.



Richard E. Daggett

Trade is one of those 'simple' concepts which seems to invite
a profusion of definitions. For example, Karl Polanyi (1957:266)
speaks of it as ". the mutual appropriative movement of goods
between hands"; Colin Renfrew (1969:152) says it is the re-
ciprocal traffic, exchange, or movement of materials or goods through
peaceful human agency"; and C.C. Lamberg-Karlovsky (1972:222) des-
cribes it as ". the reciprocal traffic of materials or goods dir-
ected by human agency from one place and or individual to another."

These definitions of trade have a number of shared elements;
that trade is a process; that there is a reciprocal physical dis-
placement of goods; that humans conduct this displacement; and that
at least two people must be involved. Renfrew alone emphasizes the
need for peace but it would seem that the act of reciprocity itself
implies the satisfaction of the needs of both parties. Certainly,
the continuation of a reciprocal relationship implies limits on or
control of threatening behavior.

Where the desire to trade exists, even in the presence of
threatening behavior, certain social mechanisms may be employed to
reduce the tension and permit the act of exchange. For example,
silent trade, or trade without face-to-face contact, has often been
employed. Acdording to John A. Price (1967), such trade is always
intersocietal, always occurs between parties at different levels of
cultural evolution and always occurs where there is reason to fear
face-to-face contact. Price provides a number of good examples to
demonstrate both the time depth and trans-cultural breadth of this
social mechanism.

According to Price, silent trade between members of bands and
tribes or chiefdoms occurs with some frequency. Unlike 'ritual
trade' which seems best to characterize silent trade at the band
and tribe or chiefdom level, silent trade between tribes or chief-
domes and states is best described as 'barter'. Moreover, it should
be noted that silent trade between bands and tribes or chiefdoms
normally takes place over a relatively short distance; such trade
between the tribes or chiefdoms and states, however, is more aptly
described as long distance trade. The practice of selecting neu-
tral sites or ports of trade is utilized to diminish fear or mis-
trust among potential trading partners. Such places guarantee the
safety of the parties, much like the concept of sanctuary.

Thus trade can be viewed as a process involving the reciprocal
movement of goods through controlled or peaceful human agency. The
unit of discussion ranges from individuals to state level societies
and, while it takes at least two individuals to conduct an exchange,
the transaction may be studied, at least partially, from the vantage
point of one of the trading partners. As a behavior, then, trade is
a response to a perceived need for an essential non-local resource.
Furthermore, trade may be viewed as a multipurpose activity with a



variety of needs being met, consciously or not, in the act of

While trade is certainly one solution to provide for the acqui-
sition of desired but locally unavailable resources, there are other
options that may also be selected by an individual group. The sub-
stitution of a local resource is one such alternative. A second
alternative, which has been utilized throughout man's existence,
is migration simply moving to a new unoccupied or underoccupied
location which better suits the needs of the mover. A third alter-
native is to make more efficient use of resources and thereby reduce
the dependence on non-local sources of supply. Other options are to
steal, or in less perjorative terms, to take what is needed or to
invent a substitute for the desired product or good.

From a cost-benefit approach, what may be the advantages and
disadvantages of selecting trade as a means to obtain desired non-
local resources? Certainly the most obvious advantage is that it
fulfills the acquisition demand; the desired good or an acceptable
alternative is obtained at a reasonable cost. What are some other
less obvious advantages?

Trade serves to promote the peace and allows the participants
to form, encourage and reinforce alliances. Another possible advan-
tage of trade is its ability to assist in the redistribution of sub-
sistence materials during periods of stress. An example of this is
provided during the nineteenth century when the Comanche, Navajo, and
Pueblo Indians participated in a three-cornered trade network; the
first group being primarily hunters and gatherers, the second basically
herders and the latter essentially agriculturists. In time of famine,
contacts, which were kept open by ritual trade, permitted the Pueblos
in particular access to the resources of their more fortunate part-
ners (Adamsl974).

Other potential advantages of trade are its ability under cer-
tain circumstances to contribute toward the preservation of a group's
internal security and the solidification of group identity where the
selection of trade is a response to unavailable but desired resources.
Finally, and perhaps most importantly, trade is integrally related to
the flow of information. The act of exchange always contains the
potential for information exchange. Hence, trade is an important
agent of communication. "It ties distant areas together and provides
a mechanism by which new ideas and subsistence techniques could be
diffused" (Wright 1974:28).

Having concluded this discussion of the potential advantages in
the selection of trade as a behavioral response, it is important to
balance the ledger by reviewing the potential disadvantages of such
a selection.

Exposure to communicable disease is certainly a clear cut dis-
advantage if the health and persistence of the individual and or group
is a measure of the success of the selected behavior. Potentially
just as deadly is exposure to foreign technology and or ideas, espe-
cially tools and practices of war which can quickly change traditional



warfare patterns into remarkably more efficient and deadly ones.
Yet another potential disadvantage related to trade is an unhealthy
reliance on external sources for critical products. Finally, it may
be stated that trade-encouraged social stratification may be only
partially beneficial. Certainly to those at the apex of a strati-
fied society the benefits are real, but to those at the bottom a
different perspective disclosed the limits of the benefits.

Trade may be studied in a variety of ways. On the basis of the
items being traded we may discern two broad categories: natural com-
modities whose natural distribution is limited, and products which
through superior technology or other economic factors are most effi-
ciently produced in a defined area (Renfrew 1969). A second approach
to the study of trade is to delineate the particular social mechanism
facilitating the exchange. Gift trade, which links partners in rela-
tionships of reciprocity, seem best able to describe the simplest (?)
and evolutionally earliest trading activities.

Long-distance trade, which has been demonstrated archeologically
to have considerable antiquity, became increasingly fraught with dan-
ger as traffic in elite goods began to predominate. The port of
trade, an example of administered trade, was a solution to the increas-
ing dangers involved in interstate trade. "The archaic syndrome com-
prised a trade carried on at set prices, and by other administrative
means. Native inhabitants provided organs for mediation and account-
ancy, while competition was avoided as a mode of transaction. .
(Polanyi 1963:30-31). Ordinarily, ports of trade are located on
coastal or riverpine plains where transport was made easier by in-
lets and lagoons. However, some were located far inland on the border
of two ecological zones, such as highland and plain, but particularly
on the border of the desert, the alter ego of the sea.

The most advanced social mechanism of trade yet devised is market
trade. A complex integration of all the participants, willingly or
not, best characterizes this category. Theoretically, open competi-
tion and the law of supply and demand function to serve the needs of
all involved.

Colin Renfrew (1975) provides an excellent review of trade as
distinguished by the location at which the event takes place. Renfrew
confesses, ultimately, to having presented an evolutionary scheme for
trade-related sites, an approach with which this writer is not in
total agreement.

Exchange may take place at the home base of one of the partici-
pants or at the common boundary of the respective territories of
exchange partners. If exchange takes place at the common boundary
it is referred to as 'boundary reciprocity'. 'Central place redis-
tribution' operates when a central person or group in a defined locale
accepts trade wares from a multiplicity of sources and facilitates
the redistribution of the various goods to the individual participants
The priesthood and temples of early complex societies in both the Old
and New Worlds may serve as an example. 'Middleman trading' and
trade which occurs between groups who choose to trade with one
another at a central place outside the jurisdiction of any one of



of any one of them, called 'port of trade', are other mechanisms of

A number of other factors, up to now left untouched, should be
discussed because of their importance in the decision of whether to
trade or not. Distance is a crucial factor, particularly as it is
primarily responsible for the decision not to trade for specific
products or services. Related to this is the presence of geogra-
phical barriers which tend to compound the distance factor. Moun-
tains, bodies of water, swamps, deserts, and thick and foreboding
forests are the more obvious examples. One should be cautious,
however, not to ascribe too much to geographical deterrents. Donald
W. Lathrap (1973), for example, argues against the notion that the
relative homogeneity of the tropical Amazon Basin stifled the develop-
ment of long-distance trade networks. Instead, by conceptualizing
the Amazon lowlands as a riverine environment, he is able to develop
an interesting argument for an early and well-developed long-distance
trade network knit by canoe travel. His extension of this network
to include the Andes and the desert coastal regions to the west,
based on archeological finds of tropical forest products in these
regions, further underscores the need for caution in viewing geo-
graphical factors as strong deterrents to human interaction.

Natural barriers have their counterparts in social barriers
which likewise have some bearing on the decision to trade or not.
Social or religious sanctions may prescribe total or partial abstin-
ence of particular goods, thereby effectively limiting the demand
for them. Likewise, unfriendly groups which occupy territory separa-
ting potential trading partners places stress and,hence,increased cost
on the act of product and or resource acquisition. Circumvention or
armed escorts are two effective but costly strategies which may be
utilized. The availability of beasts of burden and the degree of
technological achievement in transportation also play important roles
in the relative costs of conducting trade. The amount of time and
or labor an individual or group has available for investment in trade
must be taken into account. Certainly there are gradations in the
requirements of each which range from low to high depending upon the
particular circumstances of the proposed transaction. A host of
factors, therefore, must be considered in conjunction with the speci-
fic needs of an individual or group and the viability of alternative
strategies before the decision to trade or not can be made.

Thus far, it does not seem likely that a study of trade will
prove useful in the development of predictive models for human be-
havior. Too many factors weighted by particular circumstances com-
plicate the decision-making process. To make matters worse, the
actual perceptions of the decision-maker can never be known with
confidence, especially in the archeological context. However, trade
does permit a degree of understanding of the social ordering of pre-
historic groups that has been, up to now, a precarious endeavor.

Since the close of the Second World War, anthropology in general
and archeology in particular have been characterized by a rapid ex-
pansion both in terms of professional numbers and variegated inter-
ests. A proliferation of scientific technology similarly character-



izes this post-war era. The discovery of the principles of radio-
carbon analysis and its application to prehistoric chronologies
revolutionized the practice of archeology. The decades that have
followed have witnessed a veritable torrent of new dating techni-
ques and refined tools for the analysis of prehistoric human behav-
ior. To such an extent has archeology relied upon borrowed techni-
ques and theoretical break-throughs that it has earned the reputation
of being a polyglot science. That much of the borrowing has been
accomplished in an uncritical manner obviates the negative side of
this nomination.

Specialized procedures which permit the retrieval of specific
prehistoric information has progressed to the point that the actual
use of and or provenience of an array of cultural materials is pos-
sible. A brief listing of some of the available techniques and the
materials they successfully test for follows: petrological examina-
tion of thin section stone tools and pottery; optical spectroscopy -
pottery, metals, obsidian, and faience; x-ray flouresence metals,
obsidian, glass, and jade; neutron activation metals, pottery and
obsidian; infra-red absorption amber; cathode luminescence marble;
x-ray diffraction jade and emery; mass spectormeter determination
of isotopic composition lead; and fission-track dating obsidian.

Archeologically, trade has proven to be a useful investigative
tool in unraveling the organization of prehistoric societies. Be-
cause we are now able to derive the specific proveniences of certain
natural objects, we should be able to begin to separate with more
confidence potential long and short distance trade items. Goods
traded over a long distance are costlier in terms of the time and
energy required to obtain them than would be the case for locally
obtained resources. It would, therefore, seem reasonable to assume
that at least one of the trading partners of a demonstrated long-
distance trade transaction places a high value on the acquisition of
the other's trade ware.

Characteristics that would seem likely to be associated with
objects of long distance trade would be: light weight, small bulk,
low perishability, high durability, and high value. The latter cate-
gory (high value) may be further subdivided in terms of health (salt
for diet or preservation of goods; sulphur for medicinal purposes),
technology (obsidian for tools or metal for tools), or status or
idealogy (gold, jade, spondylus shell and the like used for ritual
burial, elite consumption or manufactured items for export). Stated
in another way, "Among the relevant factors governing the distance
a given object travels, are mean distance transported between exchange
transactions (in prestige exchange, it is expected that the dis-
tances are greater); transportability (expressed as a ratio of value
to weight and breakage rate in transport); and effective life, con-
sidering frequency of use, breakage rate in use, reuse-discard after
breakage, loss-recovery rate and deliberate burial (Renfrew 1977:77).

The prediction of specific types of trade-related sites should
now be within the grasp of investigators. For example, central places
should be located near unique sources of critical resources. If one
knows that a raw material is being imported from a specific area and



that such an item is not amenable to long-distance trade in its nat-
ural form, one can predict that there will be one or more intermedi-
ary stations where the resource will be converted into a more accept-
able form.

Is this the extent of what can be expected from the study of trade,
or can even more be gleaned from the prehistoric record? It would
appear that a new direction for trade studies is in the field of infor-
mation exchange. "A fundamentally new direction has been set, in which
patterns in the diffusion of archeologically recognizable commodities
are no longer the end of study but one of a battery of means by which
we seek to identify trade as a complex interactive, adaptive, and in
part, consciously directed process" (Adams 1975:452).

Henry Wright and Melinda Zeder (1977) have simulated a linear
exchange system under equilibrium conditions. They wanted to test
the idea that the extensive movement of ritual artifacts might serve
to regulate exchange systems themselves. Likewise, they wanted to
test the provocative notion that population changes in some way affect
the operation of these same trade networks. The simulation, they con-
cluded, ". .has not demonstrated that population changes determine
exchange flows, nor that some commodities are symbols regulating pro-
duction. However, the exercise has demonstrated that these things
are possible given relatively parsimonious assumptions about exchange"
(Wright and Zeder, 1977:239).

In conclusion, why trade? What advantages does it accrue that
other alternatives do not? Principally, trade always permits a degree
of communication or exchange of information. Related to this is the
fact that it always takes place in an atmosphere of peace or of con-
trolled hostility. It serves, therefore, to reduce external social
disruption while permitting access to persistence-amplifying social
mechanisms, e.g., marriage partners or assistance in time of environ-
mental or social stress. By keeping open channels of communication,
trade, principally a short-term solution to outstanding resource de-
ficiency, becomes a long-term solution to predictable environmental
and social stress.

Trade in the Bahamas

With but meager evidence suggesting a preceramic occupation of
the archipelago, there is general agreement that settlement of the
Bahamas began late in the first millennium A.D. and that colonization
progressed south to north from a source or sources in the Greater
Antilles (Granberry 1956; Hoffman 1970; MacLaury 1970; Sears and
Sullivan 1978). Julian Granberry's seminal paper on the culture
history of the Bahamas provided a tripartite, south to north, division
of the archipelago based on stylistic attributes. Recently, William
H. Sears and Shaun D. Sullivan (1978) have refined this initial div-
ision utilizing an ecological framework while essentially retaining
the original nominations. Thus the Bahamas can be divided into
southern, central and northern areas.



Ceramic and lithic artifacts share the spotlight in Bahamian
research in part because of their visible nature, and in part, due
to their value as time markers. Both classes of artifacts are char-
acterized by their durability and plasticity, characteristics which
have proven invaluable in establishing the temporal and spatial par-
ameters of essentially unknown cultures. The fortuitous consequence
of the natural deficiency in the Bahamas of certain stone types has
permitted the gross division of ceramic and lithic artifacts into
local and extra-local categories. Human agency, through the process
of exchange, is viewed as the most logical explanation for this local
deposition of extra-local products and resources (Granberry 1956;
Hoffman 1970; MacLaury 1970; Sears and Sullivan 1978).

The proposed south to north colonization of the Bahamas is appar-
ently supported by a distribution of extra-local artifacts which is
heavily weighted toward the southern Bahamas. Sears and Sullivan
(1978) report that up to 10% of the ceramic assemblage of the south-
ern islands is tempered with something other than shell- usually
quartz-sand which is unavailable in the Bahamas. Likewise, the
lithic assemblage is characterized by imported stone zemis or spirit
fetishes, monolithic axes and petaloid celts, objects typical of the
Taino culture on mainland Haiti (Granberry 1956:133). This compares
with the central Bahamas where but 1 to 2% of the ceramics were
tempered with quartz-sand and where there is a significant reduction
in the number of stone zemis, monolithic axes and petaloid celts
present (Hoffman 1970; MacLaury 1970). In the north, finally,
these artifacts are torally absent and shell tempering character-
istizes the total ceramic assemblage.

What conclusion can we draw from the foregoing regarding the
process of trade in the prehistoric Bahamas? Assuming an initial
southern colonization of the Bahamas, we can characterize the impor-
tation of extra-local products and resources as long-distance trade.
Certainly the sea can be viewed as a spatial barrier to human inter-
action, however, the simple fact of colonization implies the use of
water transport, thereby reducing the sense of barrier to one of de-
gree or perhaps convenience.

The lack of certain natural resources in the Bahamas is a geo-
logical fact. The presence of extra-local resources and artifacts
in the Bahamas is an archeological fact. We can state with some
confidence, therefore, that the presence of these non-local materials
implies a desire on the part of local inhabitants for natural re-
sources unavailable locally and or of products of superior technology.

This certainly begs the question of what was traded in return.
Sears and Sullivan (1978) offer the suggestion that crystalline salt
and dried conch, historically the two most common exports from the
Caicos to Hispaniola, were traded prehistorically to the inhabitants
of the Greater Antilles. While the Bahamas are naturally rich in
these resources, a prehistoric demand for them has yet to be estab-
lished. While salt and dried conch may be viewed as accommodating
most characteristics of items of long-distance trade (light weight,
small bulk, low perishability and high durability) it is questionable
whether they meet the standard of high value. We need to assure our-


selves that there was in fact a demand for these resources a
demand based either on local resource deficiency or high local
consumption or both.

An alternative possibility is suggested by an historical ac-
count. In his Book of the First Navigation and Discovery of these
Indies, Cristobal Colon frequently referred to being greeted by
natives offering to trade skeins of spun cotton, parrots and what
he refers to as darts (Wolper 1964). The enthusiastic reception
given Columbus and his crew certainly negates the idea that silent
trade was practiced by these natives. Trade was clearly open and
friendly. Likewise, the consistency in the objects offered for ex-
change suggests an established trade network with accepted products
of exchange. Cotton in particular should be investigated as a pos-
sible long-standing item of prehistoric trade an item which sat-
isfies the requirements of light weight, small bulk, low perishabil-
ity and very likely high value. We can speak with confidence of the
high value given to cotton by the civilizations of both Mesoamerica
and Peru. It would not be surprising to learn that Caribbean groups
also vaued this item.

Having established that trade did indeed take place in this part
of the Caribbean, how may we best describe the act of exchange?
Again, Columbus's journal offers a clue (Wolper 1964). Only three
days after initial contact with the natives of San Salvador, Columbus
reports meeting a lone Indian propelling a canoe from Santa Maria
to Fernandina. That his cargo included a string of glass beads con-
vinced Columbus that the Indian had originally embarked from San Sal-
vador. This is a classic example of the repetition of home base
trade or down-the-line trade. I suspect the trade between the sou-
thern Bahamas and the Greater Antilles was conducted in a like manner.
No evidence has yet been offered which supports the notion of the
development of central place redistribution in the Bahamas.

In summary, an understanding of trade in respect to the Bahamas
can provide an insight into important avenues of the island's culture

References Cited

Adams, Robert McC.
1974 Anthropological Perspectives on Ancient Trade.
Current Anthropology. 15(3): 239-257.

1975 The Emerging Place of Trade in Civilizational Studies.
In Ancient Civilizations and Trade, Jeremy A. Sabloff
and C.C. Lamberg-Karlovsky, eds., Albuquerque: Univ-
ersity of New Mexico Press, pp. 451-465.

Granberry, Julian
1956 The Cultural Position of the Bahamas in Caribbean
Archaeology. American Antiquity. 22(2): 128-134.





Charles A. Jr.
The Palmetto Grove Site on San Salvador, Bahamas.
Contributions of the Florida State Museum, Social
Sciences Number 15, pp. 1-26. Gainesville.

Lamberg-Karlovsky, C.C.
1972 Trade Mechanisms in Indus-Mesopotamian Interrela-
tions. American Journal of Oriental Society. 92:


Donald W.
The Antiquity and Importance of Long-Distance Trade
Relationships in the Moist Tropics of Pre-Columbian
South America. World Archaeology. 5: 170-186.

Mac Laury,

James C.
Archaeological Investigations on Cat Island, Bahamas.
Contributions of the Florida State University, Social
Sciences Number 16, pp. 27-50. Gainesville.

Polanyi, Karl
1957 The Economy as Instituted Process, in Trade and
Market in the Early Empires: Economies in History
and Theory, Karl Polanyi, Conrad M. Arensberg and
Harry W. Pearson, eds., New York: The Free Press,
pp. 243-370.


Ports of Trade in Early Societies. Journal of
Economic History. 23: 30-45.

Price, John A.
1967 Conditions in the Development of Silent Trade. The
Kroeber Anthropological Society Papers, no. 36: 67-79.

Renfrew, Colin
1969 Trade and Culture Process in European Prehistory.
Current Anthropology 10(2-3): 151-169.



Trade as Action at a Distance: questions of inte-
gration and communication. In Ancient Civilizations
and Trade. Jeremy A. Sabloff and C.C. Lamberg-
Karlovsky, eds. Albuquerque: University of New
Mexico Press, pp. 3-59.

Alternative Models for Exchange and Spatial Distri-
bution. In Exchange Systems in Prehistory. Timothy
K. Earle and J.E. Ericson, eds. New York: Academic
Press, pp. 71-90.

Sears, William H. and Shaun D. Sullivan
1978 Bahamas Prehistory. American Antiquity 43(1): 3-25.

Wolper, Ruth G. Durlacher
1964 A New Theory Identifying the Locale of Columbus's
Light, Landfall and Landing. Smithsonian Miscellaneous
Collections, Vol. 148, No. 1. Smithsonian Institution,
-.T- U- 4I ***- j-.&. __ n 0



Stephen D. Glazier

This paper will examine the ethnography of Trinidad as taken
from the scanty primary sources of the 16th century, with attention
to recent archeological findings (Harris 1976) and linguistic analy-
ses (Taylor 1977).

The current view of aboriginal Trinidad around 1595 postulates
five territorial divisions (Figure 1) probably equivalent to the chief-
tainships of the Greater Antilles" (Rouse 1948:546). The Caribs were
not present (Bullbrook 1974:4) and warlike Arawaks related to the
Igneri of the Lesser Antilles are thought to have inhabited the
island (Taylor and Rouse 1955:144), somehow spared by the earlier
Callinago migration northward.

There is something disturbing about this in terms of the ethno-
graphic record. First of all, Dudley (1899), Raleigh (1848) and
Sparrey (1906) all report Caribs in Trinidad. Also, there is no
evidence for chieftainships involving coercive power or alienation
of wealth (such as the case of those in the Greater Antilles), and
so-called Arawaks of Trinidad were actually Lokono from the main-
land (Barbudo 1874:229; Taylor 1977:2).

A tendency to regard Trinidad as part of a West Indian or
Circum-Caribbean cultural area is behind these misrepresentations.
In fact, it remains to be seen whether any connections existed with
the Greater Antilles at all. Borde (1876 1:58) appears to have
comprehended the situation correctly over a century ago when he
observed that every ethnic group on Trinidad was represented also
on the mainland. This is given additional support by the early
chroniclers (Raleigh 1848:86; Espinosa 1942:75) and by Steward
(1947:50) who includes Guiana, Trinidad, the Orinoco delta, and the
middle Amazon as one cultural area united by trade.

Archeological evidence lends credence to this position. Loven
(1935:26), in his seminal study, notes considerable migration from
Trinidad to the mainland, and the diffusion of pottery styles has
been suggested by Osgood (1942), Bullbrook (1953), and Rouse (1947,
1951, 1953). There is some archeological evidence for contact with
the mainland even during the preceramic period. Harris (1976) has
identified greenstone from the Guianas in Trinidad.

There is also linguistic evidence for this assertion consist-
ing of word-lists gathered by early chroniclers in Trinidad and the
Guianas. Unfortunately, their fragmentary nature precludes rigorous
application of lexico-statistical techniques (Taylor 1977). There
are only ten items of vocabulary for the Nepuyo of Trinidad, for
example, and fifty-two for the Yao (De Laet 1640; cf. Taylor 1977:
16-17). Dudley's word-list for the Arawak of Trinidad is by far the
most complete, and it does enable us to establish some connection
between Arawaks of Trinidad and those of the Guianas. Sixty-seven
items in Dudley's list match items collected on the mainland.


Punta de
:- /la Galera

.w f



The Serpent's Mouth or
Boca de la Sierpe

Figure 1. Trinidad

_ ___ __









I `
I /


Early traffic between the Orinoco delta and Trinidad was noted
by Raleigh (1848:85), who claims to have first heard the tale of El
Dorado from a Guianese (Lokono?) Indian who was living in Trinidad,
and later by Capuchin missionaries in the 18th century who expressed
fears that Warao might cross from the delta and interfere with their
work (Rionegro 1918, 11:216,222). More recently, Johannes Wilbert
(1977:39) claims to have made personal acquaintance with Warao who
until thirty years ago undertook trading expeditions across the Gulf
of Paria to Trinidad for purposes of trade.

It is apparent that some Guianese groups considered Trinidad
as part of their cultural sphere. This illustrates Rouse's suggestion
that bodies of water unite and structure areas in the West Indies
rather than separate them, and that the Gulf of Paria may have been
pivotal to cultural developments on Trinidad and the Adjacent main-

The Indians of Trinidad lived in large, bell-shaped communal
houses (Rouse 1948:546) that probably accommodated between a dozen
and three score persons each. The internal organization of their
households could have ranged from the bilateral extended families of
the contemporary Panare to the ad hoc political groupings of the con-
temporary Piaroa (Kaplan 1975). It is conceivable that in some places
several of these may have been found in a single settlement under a
"chief," but a fairly even scattering of single communal houses over
the countryside (each largely independent politically) was probably
the most common form of organization.

It is safe to assume that Castellano's (1850:87-88) versified
account of two powerful chieftains ruling as many provinces in 1530
is highly exaggerated. The reports of Fernandez de Oviedo (1957 II:
387) and Caulin (1966, 1:209) leave a vivid impression of loose con-
federations of very short duration. "Chiefs" probably did not have
much coercive power and the social structure appears to have been
quite egalitarian (cf. Herrera 1934, X:88).

Techniques of warfare were not militarily effective (at least
when compared with the phalanxes and counter marches of Taino warfare)
The Indians of Trinidad used wooden shields, bows and arrows, spear
throwers, wooden swords, and other military paraphernalia common also
to the Guianas.

An interesting aspect of warfare is that an alliance system
vaguely reminiscent of Marshall Sahlins' segmentaryy lineage" seems
to have been present. The analogy is not complete, however, because
while there is some evidence for segmentation and a somewhat preda-
tory expansion, there were no lineages. We may assume that the Indian
of Trinidad, like most Guianese tribes, were bilateral with shallow
genealogies to allow for the transformation of ad hoc "fictive" rela-
tionships into more legitimate ones. Though the social mechanisms of
war may have been similar to those described by Sahlins (1961), the
motivation or cognitive rationale could not, in this case, have been



In the Arawak and Carib world (as among the contemporary Piaroa)
the highest kinship unit was the group itself. Although each village
was an independent political unit, there was a belief that all "Ara-
waks" should help all other "Arawaks" especially against non-Arawaks
and above all against Caribs. However, there were also frequent alli-
ances cross-cutting "tribal" lines; for example, most of the villages
in Trinidad (regardless of "tribe") seem to have united against the
Spanish under Sedeno and others (Scott, 1925). Some early chroni-
clers found this confusing (cf. Navarrette, 1974).

This confusion is also noted in the work of contemporary scholars.
It is apparent in the recent contributions of Bullbrook (1960) and
Newson (1976). In some cases, the terms "Arawak" and "Carib" are
used with reference to two broad-ranging South American linguistic
families, and at other times with reference to specific tribes. It
is not surprising, in this regard, that linguistic affiliations did
not always serve as a basis for alliances. The various tribes, lack-
ing linguistic training, probably would not have recognized their
broader linguistic affiliations. Several so-called "Arawakan" tongues
in Trinidad were mutually unintelligible. Communication would have
been difficult without a translator. Fortunately, bilingualism was
fairly common (Laurence 1975:125) and translators were not hard to
find. Language difficulties do not seem to have posed a serious bar-
rier to trade in the area.

Raleigh reports five distinct tribal groups on Trinidad in 1595:
the Iaio (=Yao, a Carib-speaking group), Arwaca (=Lokono or True
Arawak), Salvaios (?), Nepuyos (=Sepios, another Carib-speaking group),
and the Carinepagotos (a Carib speaking group). He implies that there
may have been more such "nations" on Trinidad (Raleigh 1848:4-5), but
Sparrey (1906:301) lists these five alone. Bore (1876 1:38) equates
Raleigh's Salvaois with the Saliva, whereas Taylor (1977:16) suggests
an Arawakan affiliation. Most importantly, Harcourt (1928:86) reports
that all five of these groups also had settlements in the Guianas.

Harcourt insists that the Caribs were the original inhabitants
of the Guianas and the Arawaks, Yaos, Nepuyos and Salivaios drove
them from their homes in Trinidad and along the borders of the Orinoco.
The Carib were said to be in a constant state of warfare with these
other tribes and it was for that reason, according to Harcourt (1928:
29) that they desired protection and made an alliance with the Crown.
Arawak traditions, according to Navarrete (1974), also speak of a
time when the Carib held Trinidad and the adjacent mainland. The
Arawak say that they came "peacefully" to the area, but were appalled
by Carib customs and were driven to war against them. The Arawaks
maintain that they succeeded in driving the Carib off a good part of
their original land. The exact sequence of events, however, has yet
to be demonstrated.

The nature of political organization in the Guianas area would
have precluded a rapid conquest. Tribes do not appear to have had
permanent boundaries, and various Arawak and Carib groups seem to
have alternated along the Guianas coast, including Trinidad (Rouse
1953:68). This distribution, while not conducive to empire building,



has adaptive advantages. By the various groups alternating along
the coastline and occupying settlements on opposite sides of a
river, a single tribal group was able to exploit different ecologi-
cal niches. Lathrap (1973) has pointed out the importance of long-
distance trade in Tropical Forest settings, and Murra (1975), in
his archipelago concept, has shown how different ecological niches
could be occupied by the same people, sharing portions of each area
with other ethnic groups.

Antonio de Berrio's census of the island in 1593, estimates a
population of more than 35,000 individuals (Newson 1976:31). This
is a high figure for a landlocked area (so high, in fact, that New-
son questions it), but it is not unrealistic considering marine re-
sources made available by an extensive coastline. It is possible
that no single ethnic group on the island surpassed 10,000 individuals
and if it had been otherwise, I suspect de Berrio would have so noted.

With the coming of de Berrio, Raleigh, and Dudley, aboriginal
society began to break down. After protracted (and largely unsuc-
cessful) missionary activity and extensive acculturation (cf. Newson
1976), there are no genetic survivors left today.

From the above, the following conclusions may be drawn concern-
ing Trinidad in the 16th century: First, there were various Carib-
speaking groups on the island (contra Bullbrook 1960:4). They shared
the land with Lokonos or True Arawaks and other groups. Secondly,
Circum-Caribbean culture traits are rare on Trinidad; most archeolo-
gical and linguistic connections are recognized to be with the Tropi-
cal Forest groups of the mainland. Thirdly, there was never any
real insularity in Trinidad's prehistory (contra Taylor and Rouse
1955:546). Aboriginal Trinidad can best be understood with reference
to happenings on the mainland and vice versa.


This paper has benefited from discussions with Mr. Peter O'Brien
Harris of the Trinidad and Tobago Historical Society, Mr. Alfredo
E. Figueredo of the Virgin Islands Archaeological Society, and Pro-
fessor Irving Rouse of Yale University. Of course, I alone bear full
responsibility for the ideas presented herein.

References Cited

Barbudo, Antonio
1874 Coleccfon de Documentos Ineditos Relativos al
Descrubrimiento, Conquistas y Organizacion de las
Antiguas Espa7olas de Aiferica y Oceania. Tomo XXI,
pp. 228-239.

Borde, Pierre-Gustave-Louis
1876 Histoire de l'Ile de la Trinidad sous le Gouverne-
ment Espagnol. Premiere Partie (1498-1797).
Decouverte, Conquete, et Colonisation. Paris:
Maisonneuve et Cie., Libraires-Editeurs.






John Albert
On the Excavation of a Shell Mound at Palo Seco,
Trinidad, B. W. I. Yale University Publications
in Anthropology, no. 50.

The Aborigines of Trinidad. Royal Victoria Institute
Museum, Occasional Papers, no. 2.

Castellanos, Juan de
1850 Elegias de Varones ilustres de Indias. Segunda
Edicion. Biblioteca de Autores Espanoles, Tomo

Caulin, Antonio
1966 Historia de la Nueva Andalucia Estudio Preliminar
y Edici6n Critica de Pablo Ojer. Biblioteca de la
la Academia Nacional de la Historia, 81-82.

Dudley, Robert

The Voyage of Robert Dudley. Edited by George F.
Warner. London: Hakluyt Society.

Fernandez de Oviedo y Valdes, Gonzalo
1959 Historia General y Natural de las Indias. Edicion y
Estudio Preliminar de Juan Perez de Tudela Bueso.
Biblioteca de Autores Espanoles, tomo centesimod6cimo-
septimo-centesimovigesimoprimero. 5 vols.

Glazier, Stephen D.
1978a Trade and Warfare in Protohistoric Trinidad. Pro-
ceedings of the Seventh International Congress for
the Study of Pre-Columbian Cultures of the Lesser
Antilles, pp. 279-282.


Trinidad's Indians in the Guianas. Journal of
the Virgin Islands Archaeological Society, no. 6,
pp. 54-58.

Harcourt, Robert
1928 A Relation of a Voyage to Guiana. Edited with
Introduction and Notes by Alexander Harris. London:
Hakluyt Society.

Harris, Peter O'B.
1976 The Preceramic Period in Trinidad. Proceedings of
the First Puerto Rican Symposium on Archaeology,
pp. 33-64.

Herrera y Tordesillas, Antonio de
1934-57 Historia General de los Castellanos en las Islas y
Tierrafirme del Mar Oceano. Madrid: Academia de la
Historia. 17 vols.


Kaplan, Johanna

Laet, Jan de

The Piaroa: People of the Orinoco Basin. New
York: Oxford University Press.

L'Histoire du Nouveau Monde ou Description des
Indes Occidentales. Leyden: B. and A. Elseviers.

Lathrap, Donald W.
1973 The Antiquity and Importance of Long-Distance
Trade Relationships in the Moist Tropics of Pre-
Columbian South America. World Archaeology, vol.
5, no. 2, pp. 170-186.

Laurence, K. M.

Continuity and Change in Trinidadian Toponyms.
Nieuwe West-Indische Gids, 50(2,3): 123-142.

Loven, Sven

Murra, John

Origins of the Tainan Culture, West Indies.
Goteborg: Elanders Boktryckeri Aktiebolag.

Formaciones economics y politicas del mundo
andino. Lima: Instituto de Estudios Peruanos.

Navarrete, Rodrigo de
1974 Relacion de las provinces y naciones que los yndios
llamados Aruacas. Coleccion de Documentos y Organi-
zaci'n de las Anti uas Posesiones Espagolas de
America y Oceania, Tomo XXI, pp. 221-228.

Newson, Linda A.
1976 Aboriginal and Spanish Colonial Trinidad: A Study
in Culture Contact. New York: Academic Press.

Osgood, Cornelius
1942 Prehistoric Contact Between South America and the
West Indies. Proceedings of the National Academy
of Sciences, vol. XVIII.

Raleigh, Walter
1848 The Discovery of the Large, Rich and Beautiful
Empire of Guiana. Edited by Robert H. Schomburgk.
London: Hakluyt Society.

Rionegro, Froylan de
1918 Relaciones de las Misiones de los PP. Capuchinos
en las Antiguas Provincias Espa olas Hoy Rep blica
de Venezuela, 1650-1817. Sevilla: Tip. Zarezuela/
Tip. "La Exposici6n". 2 vols.



Rouse, Benjamin Ivring
1947 Prehistory of Trinidad in Relation to Adjacent
Areas. Man, old series, 47(103): 93-98.

1948 The Ethnography of Trinidad. Handbook of South
American Indians, Julian H. Steward, Ed. Bureau
of American Ethnology: Bulletin, 143(4): 545-546.

1951 Areas and Periods of Culture in the Greater Antilles.
Southwestern Journal of Anthropology, 7: 248-265.

1953 The Circum-Caribbean Theory, An Archaeological Test.
American Anthropologist, 44(2): 188-200.

Sahlins, Marshall
1961 The Segmentary Lineage: An Organization of Preditory
Expansion. American Anthropologist, 63: 322-344.

Scott, John
1925 The Discription of Trinidada. Colonising Expeditions
to the West Indies and Guiana, 1623-1667. Edited by
V. T. Harlow. London: Hakluyt Society.

Sparrey, Francis
1906 The description of the Ile of Trinidad. Hakluytus
Posthumus. Glasgow: James MacLehose and Sons, vol.
XVI, pp. 301-302.

Steward, Julian H.
1947 Handbook of South American Indians. Bureau of Ameri-
can Ethnology, Bulletin no. 143.

Taylor, Douglas MacRae
1977 Languages of the West Indies. Baltimore: Johns
Hopkins University Press.

Taylor, Douglas MacRae and Benjamin Irving Rouse
1955 Linguistic and Archaeological Time-Depth in the West
Indies. International Journal of American Linguistics,
vol. XXI, pp. 105-115.

Vasquez de Espinosa, Antonio
1942 Compendium and Description of the West Indies. Charles
Upson ClarK, Trans. Washington, D. C.: Smithsonian.

Wilbert, Johannes
1977 Navigators of the Winter Sun. The Sea in the Pre-
Columbian World. Elizabeth P. Benson, Ed. Washington,
D.C.: Dumbarton Oaks Research Library. pp. 16-46.



Back issues of the FLORIDA ANTHROPOLOGIST, if available, .may
e purchased from the Assistant Editor. Volumes 1-30 are $6.00 per
volume ($1.50 for single and $3.00 for double numbers); Volume 31 and
ubsequent volumes are $8.00 per volume ($2.00 for single numbers).
volumes 25:2, pt. 2 (Publication No. 6), 28:3, pt. 2 (Publication NO. 7)
and 31:2, pt. 2 (Publication No. 9) are $2.00 each. Volumes. 1; 2:3-4:
:1-4; 6:1-4; 7; 8; 9; 10; 11; 12:1-3; 13:1; 16:1,3,4; 17:1; 18:1; 20:1-
1; 23:2; 24:1; 25:3-4; 26:1; 27:1,3; 29:2; and 30:1 are out of print.

monograph series, may be purchased as follows: Numbers 6, 7, and 9
at $2.00 each, Number 8 at $3.00. Numbers 1, 2, 3, 4, and 5 are out
.f print.

Orders with remittances for any of the above should be sent to
Irving Eyster, Route 1, Box 96, Islamorada, FL 33036. Please allow
six to eight weeks for delivery. The Society has no resale discount
for dealers or resale agents. Please include $1.00 for postage and
handling with each order.

Back issues may also be obtained from Johnson Reprint Corp.,
11 Fifth Avenue, New York, NY 10003. Vols. 1-13 are available for
$20.00 per volume, $10.00 per double numbers, and $4.50 per single
number. Numbers 1-5 of the PUBLICATIONS are $5.50 per number.


THE FLORIDA ANTHROPOLOGIST publishes original papers in all sub-
Eields of anthropology with an emphasis on archeology. Contributions.
from allied disciplines are acceptable when concerned with anthropolo-
gical problems. The journal's geographical scope is Florida and
adjacent regions.
Manuscripts should be .double-spaced and typed on one side only
of 8h by 11 inch paper. Authors should submit the original and one
copy of their manuscript. For matters of style and reference see a
recent issue of the journal. Text references should be similar to
*(Smith 1970:44-45; Jones 1972: P1. II, b)." Footnotes are normally
not accepted.

All illustrative material--line drawings (in India inki and
photographs--should be included in one numbered series of "Figures,."
Bach figure must be accompanied by a brief descriptive caption; parts
of figures are designated by lower case letters. All tabular material
must be included in a separately numbered series of "Tables." Authors
should examine a recent back issue of the journal for figure and table

Twenty-five reprints without covers of each article are furnished
each author or group of authors.


Full Text