“Gods … Adorned with the Embroiderer's Needle”: The Materials, Making and Meaning of a Taino Cotton Reliquary

Abstract

"A unique cotton Taíno reliquary— the only extant example currently known— provides an unprecedented window onto the complex mortuary and ritual ceremonies of the pre-Hispanic Caribbean. This study explores its cultural context as recorded by the early Spanish and French chroniclers and missionaries who were witness to the use and beliefs surrounding these objects in both the Greater and Lesser Antilles. It provides the first AMS radiocarbon date for the reliquary, placing it within a firmer historical context. It also examines the woven sculpture in some detail, providing a review of the manufacture process and a detailed study of the components— cotton, animal hair, lianas, gourd, resins and shell— that went into its creation.
From the wrapping of important cemís (representations of spirits) in cotton, to the binding of the skeletal remains of venerated ancestors within elaborate weavings, cotton had an intrinsic value as a material that wrapped and bound the ancestors to the living and the living to each other."

Full text

... and the images of gods which are adorned
with the embroiderer’s needle represent men
who were religious and just and known to have
conferred some benefit on the people, and were
numbered among their saints by the common
beautification of the priests and country. The
teeth which protrude from their mouths hang
from a human skull, which is hidden beneath
the embroidery; these images which represent
men who became saints ... used to give public
oracles to the people, but fell silent all together
at our God’s arrival in this country. —letter
from Bishop Alessandro Geraldini (b. 1455, d.
1525) to Pope Leo X, ca. 1519 or 1520 (in Sym-
cox 2002:133).
To accompany this letter, the first Bishop of
Santo Domingo sent several examples of cot-
ton figures, or cemís (representations of
ancestors, spirits, or deities), as gifts to the Pope, sug-
gesting that they be displayed in the vestibule of St
Peter’s Basilica. This is not the first mention of such
“GODS ... ADORNED WITH THE EMBROIDERER’S NEEDLE”:
THE MATERIALS, MAKING AND MEANING
OF A TAINO COTTON RELIQUARY
Joanna Ostapkowicz and Lee Newsom
A unique cotton Taíno reliquary— the only extant example currently known— provides an unprecedented window onto the
complex mortuary and ritual ceremonies of the pre-Hispanic Caribbean. This study explores its cultural context as recorded
by the early Spanish and French chroniclers and missionaries who were witness to the use and beliefs surrounding these
objects in both the Greater and Lesser Antilles. It provides the first AMS radiocarbon date for the reliquary, placing it within
a firmer historical context. It also examines the woven sculpture in some detail, providing a review of the manufacture
process and a detailed study of the components— cotton, animal hair, lianas, gourd, resins and shell— that went into its cre-
ation. From the wrapping of important cemís (representations of spirits) in cotton, to the binding of the skeletal remains of
venerated ancestors within elaborate weavings, cotton had an intrinsic value as a material that wrapped and bound the
ancestors to the living and the living to each other.
Un exclusivo relicario taíno de algodón— el único ejemplar conocido hasta la fecha— provee una mirada sin precedentes hacia
las complejas ceremonias y rituales del Caribe prehispánico. El presente estudio explora el registro cultural asociado al reli-
cario a través de los registros realizados por los primeros cronistas y misioneros españoles y franceses, quienes fueron testi-
gos de los usos y creencias que rodearon a estos objetos tanto en las Antillas Mayores como Menores. Este objeto provee una
de las primeras dataciones por radiocarbono (AMS), localizándolo dentro de un claro contexto histórico. Al mismo tiempo,
se examina la estructura del tejido en cierto detalle, brindando una revisión del proceso de manufactura y un detallado estu-
dio de los componentes— algodón, pelo animal, lianas, calabaza, resinas y conchas— utilizados para su creación. Desde la
envoltura de importantes cemís en el algodón, hasta la unión de restos de esqueletos provenientes de venerados ancestros den-
tro de elaborados tejidos, el algodón tuvo un valor intrínseco como material que “envolvió y unió” a los ancestros con los
vivos, y a los vivos entre sí.
Editor’s Note: For the first time, Latin American Antiquity is making color images available as “Supplemental Materials” linked
to the online version of this article. The supplemental online images are referenced throughout the article as Supplemental Figure
1, Supplemental Figure 4, and so on, with their numeration corresponding to figures in the print version. Color images are not
provided for Figures 2, 3, and 11.
Joanna Ostapkowicz 䡲World Museum Liverpool, William Brown Street, Liverpool, L3 8EN, UK (Joanna.Ostapkowicz@
liverpoolmuseums.org.uk)
Lee Newsom 䡲Department of Anthropology, The Pennsylvania State University, University Park, PA 16802
Latin American Antiquity 23(3), 2012, pp. 300–326
Copyright ©2012 by the Society for American Archaeology
300
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

“GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 301
sculptures in documents from the early period of
European contact, but it is the most illuminating not
only in clearly describing the presence of the skull
within the “embroidered” cloth, but also distin-
guishing these figures from other examples of Taíno
cemís, which existed in a wide variety of forms.
Geraldini clearly identifies these as representing—
and holding the remains of— behiques (shamans) or
men/women of reputation, ancestors who had sig-
nificant roles in life, and a visible and tangible role
after death. These were clearly understood as ven-
erable beings who became intercessors or mediators
on behalf of the living— analogous, as Geraldini
points out, to saints. The Europeans, with their own
deep history of reliquaries, quickly grasped the
power behind these cotton figures, and used them
as propaganda for the power of the Christian church
in “silencing” the beliefs of a newly discovered world
(see also Vallejo 1913:45).
For the Taíno, the physical manifestation of the
ancestors in this way made the past visible and
accessible— consulting them through prescribed
ritual ensured that they continued to have impact
on the lives of the living. According to Father
Ramon Pané (1999:21), the Jeronomite friar who
lived among the Taíno ca. 1494–1498, reliquaries
were one of the many categories of cemís: “All or
the majority of the people of the island of Hispan-
iola have many [cemís] of various sorts. Some con-
tain the bones of their father and mother and
relatives and ancestors.... [there are] some that
speak, and others that cause the things they eat to
grow, and others that make it rain, and others that
make the winds blow.” Las Casas (1967:I:633)
noted that certain reliquaries were said to hold the
bones of caciques and were named after them.
These sculptures appear to be an elaboration of the
more common Taíno practice of keeping ancestral
remains in baskets or gourds suspended from the
rafters of communal houses (Colón 1992:75), and
the practice may have emerged as cacical power
grew after ca. A.D. 600 (Curet and Oliver 1998;
Ostapkowicz et al. 2011). These larger reliquaries
were consulted as oracles and kept in isolated
places— cacical houses segregated from the vil-
lage, or in caves; access to them was restricted,
although their pronouncements were ostensibly for
the benefit of the community. Some cronistas
(chroniclers) believed that caciques used reliquar-
ies for their own advantage, manipulating their
“words” for their own benefit (Las Casas
1967:I:633; Oviedo 1992:I:112). Regardless of the
politics that may have underscored their use, they
clearly had an important role in maintaining the
genealogical histories and connections between the
living and their ancestors (Oliver 2009:251). As
such, it is unlikely that they were given voluntar-
ily to the Spanish, like many other Taíno valuables
during the early exchanges— but rather had inalien-
able qualities that tied them intimately to their
communities.
This paper explores the context and meaning of
textile reliquaries through a review of cronista doc-
umentation and an examination of the only known
surviving cotton example, held in the collections
of the Museum of Anthropology and Ethnography,
Turin, Italy (MAET) (Figure 1, Supplemental Fig-
ure 1). The focus is on how the materials used in
its construction were brought together to create the
final form, and how each offers insights not only
into the considerable labor invested in the creation
of the cemí, but its layering of value and meaning.
Cotton Cemís: Cronista Documentation
There are several early accounts that describe cot-
ton reliquaries, including Andrés Bernaldez’s brief
mention of cotton figures that Columbus himself
brought to his rectory in 1496, shortly after his
return from the second voyage (Loven 1979:585).
Bernaldez noted that “the Admiral brought at that
time many things that the Indians used: crowns,
masks, belts, collars and many things woven from
cotton, the devil figuring in all of them..., some
sculptured in wood, some made of the same cot-
ton in bulk and some were jewelry.” The repeated
references to cotton makes clear the importance of
this material as an indigenous valuable woven not
only into status items such as belts, but also into
sculptures, to which the Spaniards reacted with a
mixture of both curiosity and repulsion.
Shortly after the turn of the sixteenth century,
Martyr D’Anghera (1970:167) sent four “idols to
whom the islanders pay public worship... in the forms
of seated figures, out of plaited cotton, tightly stuffed
inside” to Cardinal Ludovico D’Aragón. He identi-
fied these figures as cemís (or zemis/zemes), a term
he understood to include different types of repre-
sentations, and noted that the Taíno “believe the
zemes to be intermediaries between them and God....
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

each cacique has his zemes, which he honours with
particular care” (Martyr D’Anghera 1970:167).
The circulation of these figures in Europe was
to continue well into the mid-seventeenth century,
passing through some of the most influential houses
in Europe. Only glimpses of these exchanges
remain in surviving archival records. Fr. Francisco
Ruiz, one of the first Franciscans sent to Hispan-
iola, returned to Spain ca. 1501–1502 laden with
“curios”—including enough “idols of diverse man-
ner” to fill one or two chests, which he gave to the
archbishop of Toledo, Cardinal Francisco Jiménez
de Cisneros (d. 1517) (Vallejo 1913:45; see also
Alegría 1981:60; Royo Guardia 1947:151). Among
these were “frightening forms of malignant spirits,
with eyes and teeth made of fish bones and
[padded?] bodies of cloth... and feet and ears of cot-
ton, all made by [Native] hands” (Vallejo 1913:45).
This collection of idols was reportedly still housed
at the University of Alcalá in 1547 (Alegría
1980:434), but sometime after this it was widely
dispersed, with at least one piece reaching the
famed collections of the Munich Kunstkammer.
This collection, largely amassed by Albrecht V,
Duke of Bavaria (1550–1579), was inventoried in
1598, listing a cotton cemí with “small white and
red interlocking rings... [and]... big eyes of blue
glass” linked to Cardinal Cisneros’ collections
(Feest 1991:581). By 1626, the cemí was illustrated
in Lorenzo Pignoria’s Imagini de gli dei Indiani,
where it was noted to have come with Spanish tes-
timonies to it speaking (Pignoria 1626:563–564;
see also Feest 1986:191), echoing Geraldini’s
description of similar pieces over a century earlier.
The use of blue glass demonstrates the early incor-
poration of European goods in traditional Taíno
objects, recalling the blue glass beads woven into
the cotton cemí currently in the collections of the
Pigorini Museum in Rome (Taylor et al. 1997). 1
The Taíno suffered almost complete cultural
dissolution by the mid-sixteenth century due to the
spread of epidemic diseases, imposed slavery, war-
fare and violence resulting from European invasion
and colonization of the Greater Antilles, and the
manufacture of such figures likely ceased at this
time. However, their Carib/Kalinago neighbors in
the Lesser Antilles continued their traditional prac-
tices largely undisturbed well into the seventeenth
century, and these practices included the use of
similar cotton figures that were said to “give ora-
cles” (Rochefort 1666:280). Charles de Rochefort
(b. 1605, d. 1683) (1666:284) noted that “they
expect... the sentence of their life or death from
those detestable oracles, which they receive by the
means of those puppets of cotton, wherein they
wrap up the worm-eaten bones of some wretched
carcass taken out of the grave.” These oracles were
appeased by numerous services: “they not only
consecrate to them the first of their fruits, but they
also devote to them the most sumptuous tables of
their feasts; they cover them with the most delicate
of their meats, and the most delicious of their
drinks; they consult them in their affairs of great-
est importance, and are govern’d by their wicked
counsels” (Rochefort 1666:284). The reverence
shown to these figures by the Carib/Kalinago and
their important roles in everyday life parallel what
had been documented for the Taíno over a century
earlier. This was foreign to Rochefort and many
of his contemporaries, who documented the prac-
tices surrounding reliquaries through the prism of
Eurocentric bias.
One of the most detailed accounts comes from
Rochefort’s contemporary, Jean-Baptiste du
Tertre (b. 1610, d. 1687), a missionary who
worked in Guadeloupe and Martinique. He noted
that “the devils sometimes nest in bones taken
from the sepulcher and wrapped in cotton, return-
ing oracles when questioned... it is the heart of
death that speaks” (du Tertre 1667:369). These
were used in various forms of divination, includ-
ing one ceremony involving their immersion in
water: according to du Tertre (1667:369) they
were thrown into the sea and if they sank, it fore-
told of storms and danger; if they floated, fine
weather and auspicious times. Their power could
be manipulated against “enemies”:
They use these speaking bones to bewitch
those against whom they hold a grudge... [by
taking] what is left of the food or drinks of their
enemies, or maybe some other effects that
belong to him, then they wrap it with these
bones and we see that he looses his usual vigor,
a slow fever weakens him, takes over him and
he dies languishingly, and there is no remedy
that would help him recover his health [du
Tertre 1667:369].
Du Tertre (1667:369) goes on to recount a specific
case from Guadeloupe, where the family of a mur-
302 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

dered man sought revenge: “by stain[ing] cotton
with the blood of [the victim] and plac[ing] it with
these bones of death, and we immediately wit-
nessed [the perpetrator] wasting away, and after lan-
guishing for two years, he died.”
At some point between 1637 and 1658, du Tertre
(1667:369–370) reports that several “cotton idols
in human form, having cakes of soap seeds instead
of eyes [possibly soapberry seeds, Sapindaceae—
Vega 1987:6], and a type of helmet made of cotton
on their head” were found in caves on Martinique
by two Carib/Kalinago. They believed them to be
the “Gods of the Ygneris, whom they massacred,”
and trembled in fear when they approached them.
Orders were given to collect the figures, but the
Carib/Kalinago refused to enter the cave again—
du Tertre (1666:370) notes that this was later
accomplished in secret, without their knowledge.
The figures were then placed in a crate and sent
back to France as a gift for the Duke of Orléans. In
a twist of fate, the frigate carrying them was cap-
tured by the Spanish, and once the images were
found, the captain was charged with idolatry and
brought before the Inquisition. Although the cap-
tain was acquitted of the charges, it is unlikely that
the figures survived.
Turin Cotton Cemí: Provenance and History
The cotton cemíssent to Europe during the fifteenth
to seventeenth centuries eventually deteriorated or
were otherwise destroyed, as no further record of
them remains. The only extant example of a cotton
reliquary currently known is housed in the MAET
(Figure 1, Supplemental Figure 1). It is understood
to have been found in a cave west of Santo
Domingo, Dominican Republic, at some point prior
to 1891 (Cronau 1892:I:263; Fewkes 1891:174;
Vega 1987:1). Its preservation in a cave for over
three centuries initially appears remarkable, but is
in keeping with previous accounts of where such
objects were found (du Tertre 1666:370), and such
desiccating environments are generally conducive
to exceptional organic preservation. According to
the personal notes of Jesse Walter Fewkes
(1903:6:45a), the cemí was found by a man hunt-
ing wild boar, who was so alarmed by what he saw
that he struck out at it with a machete— but later
returned with help to remove it from the cave. It
eventually came to the collection of Sr. Cambiaso,
and it is likely here that Rudolph Cronau sketched
it for inclusion in his book Amerika (Cronau
1892:I:263) (Figure 2). Captain Nathan Appleton,
a well-connected Bostonian, also acquired an illus-
tration of the cemí by Sr. A. Rodríguez, sending it
to Fewkes at the Smithsonian (Fewkes 1891; see
also Seelye 1892:133). By 1903, when Fewkes trav-
eled to Santo Domingo to study the cemí, he learned
that Cambiaso had sent it to Italy— Genoa, accord-
ing to the family (Vega 1987:4). In his review of the
piece, Fewkes (1907:214) noted: “It is much to be
regretted that our knowledge of this figure, which
could shed so much light on the mortuary rites and
worship of the prehistoric Antilleans, is so imper-
fect. The author was told that it is now somewhere
in Italy, but whether it is lost to science could not
be learned.”
The cemí disappeared from wider knowledge
until 1970, when Bernardo Vega discovered an
archival photograph of it in the British Museum
(Figure 3), and was able to trace the reliquary to the
university museum in Turin (Vega 1987:1). This
photograph, which appears to date to the early twen-
tieth century, if not slightly earlier, shows the reli-
quary with the wooden canopied cemí also in the
MAET, suggesting that the history of the two pieces
is to some degree entwined (Fewkes illustrates both
in his 1891 publication, as does Seelye
[1892:132–133], and Vega [1987:12] suggests that
they were probably found together). Curiously, a
small tag on the cotton cemí’s right hand bears the
name “Hamilton Prints, Lot 39[?]2,” possibly an
auction label (Dan Bruce, personal communication
2009), suggesting a more complex history between
private collection and museum acquisition. It is not
as yet clear when the sculptures entered the MAET:
the earliest record— a Stampa Sera newspaper arti-
cle dated March 2, 1940—indicates that Ernesto
Schiaparelli, Director of Turin’s Egyptian Museum,
donated the cotton cemí to Giovanni Marro, the
founder and first director of the MAET, at some
point prior to 1928 (Rosa Boano, personal com-
munication 2011). The MAET opened in 1926, but
the first mention of the cotton cemí in the museum
records appears to be a catalogue entry tentatively
dated to the 1950s, where it is listed under the num-
ber 1676 as a “ Peruvian [sic] anthropomorphic idol
made of interlaced vegetable fiber, encasing a skull
and featuring one dark and one light eye” (Masali
and Pia 1991:86). There is therefore a gap between
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 303
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

304 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 1. Turin cotton cemí. A.D. 1439–1624 (95.4 percent), with the greatest probability (76.7 percent) at A.D. 1439–1522. H: 55 cm (max, on stand), W: 35.5 cm. Gossypium sp., ante-
rior human skull (including mandible), internal cane framework for arms and legs with central carved wooden support and stone base, resins, shell, gourd, pigments(?) Part of the
stand that supports the cemi is visible on the right image, at the base of the its spine. (Courtesy, The Museum of Anthropology and Ethnography, University of Turin, Italy. See
Supplemental Figure 1 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

1903, when Fewkes noted it being sent to Italy, and
ca. 1928, when the cotton cemíwas transferred from
Schiaparelli to Marro. Further archival work is
needed to clarify its twentieth-century history.
Vega (1987 [1971]) initiated the first major
study of the piece and its history, including the first
radiographs of its internal structure. These clearly
show the human cranium encased in the cotton
weave, as well as a bulky, opaque substance at the
figure’s center, which Vega suggested was stone.
His study has been followed by a growing body of
research on the cemí (AMT ca. 1975; Girotti and
Meaglia 2001; Guidi and Appendino 1973; Masali
and Pia 1991; Meaglia and Girotti 2001; Ripley
1980), most recently the CT study by Martina et
al. (2010), which has brought the internal structure
of the cemí to light in magnificent detail (Figure 4,
Supplemental Figure 4). The work undertaken here,
initiated by the first author in 2005, builds on these
foundations by focusing on its radiocarbon age,
construction materials, and process of manufacture.
Radiocarbon Results
A 75 mg sample of cotton was extracted by MAET
conservators from two separate areas of the cemí:
the left side flap at the back of the cemí’s head and
the right knee. The former, the larger of the two
samples, was submitted for radiocarbon dating,
while the smaller sample from the knee area was
analyzed by Newsom for material identification.
The radiocarbon result of 395 ± 27 B.P.
(OxA–15359; ␦13C = –22.4‰) calibrates with 95.4
percent confidence to A.D. 1439–1522 (76.7 per-
cent), 1574–1584 (1.6 percent) and 1590–1624
(17.1 percent): of these, the greatest probability
lies with the period A.D. 1439–1522. While tradi-
tional Taíno lifeways continued for some decades
after contact, it is extremely unlikely that this object
would have been made as late as 1575–1624: as
will be discussed below, such sculptures required
an intact indigenous sociopolitical system for their
meaning and relevance, and due to population
decline and forced acculturation this essential
framework disintegrated by the mid–sixteenth cen-
tury. Moreover, the piece does not feature any post-
contact materials, as seen in other large cotton
sculptures, such as the blue glass in both the Pig-
orini and Cisneros cemís, the latter in Europe by
ca. 1502. This would suggest that the cotton for the
cemí was collected, and likely spun and woven,
some decades prior to, or very shortly after, Euro-
pean contact. Cotton fibers (technically, seed hairs
or “lint”) are produced by the plant on an annual
basis, and so represent one year of growth; they are
harvested regularly, and hence are an excellent
material for radiocarbon dating, as they are not sub-
ject to in-built age. Although both raw and spun
cotton were stored in cacical “warehouses,” as well
as people’s homes, it is unlikely that they were
stored for long periods of time given the constant
demand for cotton products, such as hammocks,
naguas (skirts), and body ornaments. Hence the
date provides a good indication of when the cemí
was constructed.
Fiber Analysis
The sample from the knee area of the cemí con-
sisted of four small segments of cordage (desig-
nated TC1 through 4), each analyzed for fiber
content. The primary fiber mass is consistent in all
four samples, each being predominantly composed
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 305
Figure 2. One of the earliest published images of cotton
cemí, in Rudolph Cronau’s Amerika (1892:I:263; see also
Fewkes 1891; Seelye 1892:133). Note the absence of the
strap on the forehead and the loose fibers at the top of the
head, suggesting that damage to the back of the cemí’s
head preceeded the publication. The damage is also appar-
ent in the illustration in Fewkes’s 1891 publication.
(©British Library Board [9551.1.6].)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

306 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 3. Archival photograph of cotton and wood cemís (ca. late nineteenth/early twentieth century). Note that the strap
on the cotton cemí is now positioned immediately behind the second ridge of the forehead. (By kind permission of the
Trustees of the British Museum.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

of Z-twisted luminescent cotton (Gossypium sp.,
Malvaceae) fibers ca. 20–25 µm diameter with
clear cell lumens and walls, acuminate tips, fairly
regular to quite regular reversals in direction or
twists, and ranging in color from amber to white
or pale bluish grey (Figure 5, Supplemental Fig-
ure 5). Stellate hairs from the original involucres
of the cotton boll are also present. In general, the
fiber morphology and color variation, with the
notable exception of one stained a deep red, com-
pare well with fiber from uncultivated (naturalized
or feral) cotton from the Caribbean and South
Florida (Figure 5, Supplemental Figure 5). The
original position of the red fiber in the cotton twine
is uncertain, having first observed it under high
magnification, i.e., after having extracted a mass
of fibers from the surface and mounting them on
a glass slide for detailed analysis. If not intrusive,
the red cotton fiber likely represents deliberate use
of natural pigments and dyes as traditionally
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 307
Figure 4. Three CT views of the interior of the cemí, showing the shaped stone and wooden base supports. (Courtesy Dr
María Cristina Martina, University Institute of Diagnostic and Interventional Radiology [Director Prof. Giovanni
Gandini], San Giovanni Battista, Molinette Hospital, University of Turin. Italy; adjusted in Photoshop by Ostapkowicz.
See Supplemental Figure 4 for color image.)
Table 1. Animal Hair Associated with the Turin Cotton Cemí Samples.
Type Sample Color Diameter(microns) Medula Cuticle Ends Assignment
1 TC1 tan 8.5-10 absent coronal/ blunt bat
petaloid
2 TC4 bluish-black 10 absent coronal/ blunt bat
petaloid
3 TC4 reddish brown 34-35 unicellular, imbricate or cut primate (cf.)
regular; mosaic
uniserial ladder
4 TC2 brown 24-29 obscure mosaic- blunt; non-human
irregular frayed mammal
5 TC3 rose 30-36 obscure mosaic- frayed;cut? non-human
irregular mammal
6 TC4 yellowish 25-29 obscure mosaic- cut non-human
brown irregular mammal
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

employed in the Caribbean (e.g., achiote, Bixa
orellana, the putative source of red body paint;
[Newsom and Wing 2004].
In addition to the cotton fibers, several types of
animal hair were discerned among the cordage sam-
ples; essential details are summarized in Table 1.
Mammalian hairs are generally of two basic
types— underhair and overhair (i.e., “guard hair,”
the most useful type for taxonomic identification)
—the respective anatomical and morphological
characteristics for which vary not only down the
length of the hair shaft, but also depending on age
and other factors (Petraco and Kubic 2004; Teerink
1991). The samples in question, being mainly small
fragments of hair shaft, are difficult to classify tax-
onomically; nevertheless, assignments are made to
the extent possible based on the features preserved
in each specimen. Caribbean fauna is famously
impoverished of mammalian taxa (Hedges 2001),
a situation that was somewhat modified by prehis-
308 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 5. Turin cemí cotton fiber. Upper row: left, whole cotton cord (TC2) showing color variation (30x); center, stellate
hair (TC2) (400x); right, amber variant showing classic twist, cell walls, and lumen (400x). Center row: left, natural color
variation (TC3) (100x); right, natural color variation in modern reference sample from Big Mound Key, Charlotte
County, Florida. Bottom row: red-stained cotton fiber (TC2), left (100x), right (200x). (See Supplemental Figure 5 for
color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

toric human introductions of particular taxa (Wing
2001); thus we are able to examine representative
hair from a proportionately large sample of that
biodiversity.2
Hair types 1 and 2 are bat hair (order Chiroptera;
one specimen each) (Figure 6, Supplemental Fig-
ure 6). The second was fairly well integrated with
surface layers of the cotton cord, but whether either
or both were deliberately incorporated is uncer-
tain. They most likely represent naturally shed hairs
that incidentally became associated with the cemí
during its history in the cave, or potentially in the
raw cotton when it was stored in houses (includ-
ing storehouses) where bats may have roosted. But
the intriguing possibility that bat hair was inten-
tionally incorporated merits further investigation,
especially considering the deep symbolic impor-
tance of the bat in Taíno iconography and mythol-
ogy, including its possible links to death and opias
(spirits) (García Arévalo 1997). Parallels to its use
in textiles exist further afield: bat hair was report-
edly incorporated into the finest Peruvian weaves,
reserved for Inca elite (Murra 1962:719).
Four additional types of animal hair were
observed in association with the cotton fiber. One
of these was deeply integrated into the cotton yarn,
which had to be untwisted to remove it for closer
scrutiny; both ends appear to have been cut.
Another hair is artificially rose-colored, reminis-
cent of the red-tinted cotton fiber described above.
It was first observed loosely curled and wrapped
around the cotton cord, extending out from one
end. This configuration may suggest a true associ-
ation, with the hair having become partly dislodged
and pulled loose, and indentations along its length
(Figure 7, Supplemental Figure 7) suggest incor-
poration in a 2-ply yarn or plain weave textile. The
last two hair types were mostly external to the spun
cord, partially integrated with the surface layers of
cotton fiber; that both ends of one of these appear
cut may suggest deliberate incorporation. The pre-
cise taxonomic assignments for these four addi-
tional hair types are under further investigation,
including aDNA, to clarify their origin(s).
Construction: Preparatory Stages
The cemí is a remarkable achievement of a
weaver’s skill, transforming what is essentially a
two-dimensional textile into a three-dimensional
object. The effort of bringing the materials
together— from the gathering of the cotton bolls to
the spinning— was considerable. As Stark et al.
(1998:10) point out, “[g]rowing, picking, ginning,
carding and spinning cotton represented a sub-
stantial part of the labor in cotton textiles... These
steps plus dyeing and weaving... form a labor inten-
sive sequence that continually adds value.” It is
likely that the Taíno harvested free-ranging wild
cotton plants (Colón 1992:70; Oviedo 1992:II:14)
alongside cultivating cotton among their conucos
(agricultural fields) and home gardens (Cosa in
Olazagasti 1997:136; Newsom 2008; Sauer
1992:56). As such, harvesting was dependent on
collecting bolls in different areas, at different times
during the growth season, as not all bolls ripen at
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 309
(a) (b) (c)
Figure 6. Turin cemí bat (Chiroptera) hair Type 1 (a) and (b) (400x, 600x) and (c) Type 2 (600x). (See Supplemental
Figure 6 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

310 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
) (a
)b(
(c)
(e
)(e
f)(
f)
Figure 7. Rose-colored hair (TC3): (a) looped and crossed over itself (200x); (b) in original position (40x); (c) rupture
(400x); (d) two indentations (200x) and close-up of one (e); (f) indentation in reference sample of merino wool hair from
a textile sample (600x). (See Supplemental Figure 7 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

the same time (Langer and Hill 1991; Rehm and
Espig 1991)—especially those growing as escapees
from cultivation. This required knowledge of the
growth of cotton in various areas, and an invest-
ment of time to visit the plants regularly during the
yield season for a good supply of cotton. Once suf-
ficient quantities were gathered, processing would
start with cleaning the cotton of leaves and detri-
tus, and extracting the seed from the fiber by hand,
a laborious process that had to be done well other-
wise it would affect the quality of the yarn: poorly
prepared cotton results in lumpy yarn that breaks
easily, while spinning rates are partially dependent
on fiber quality and the work invested in fiber
preparation (Vreeland in Tidemann and Jakes
2006:301–302).
Additional materials may have been added at
this stage, whether to enhance the strength or look
of the yarn, or perhaps to enhance its symbolism.
In the case of the cotton cemí, these appear to have
included at least some animal hairs, some of which
are dyed. Dyeing fibers or hairs was yet another
processing step that required its own sequence of
preparation, involved and time consuming in its
own right— between the collecting, cleaning, dye-
ing and drying. There are only brief cronista ref-
erences to dyed materials among the Taíno, such
as the use of “canes of different colors interwoven
with the most marvelous skill” to cover house
beams near Puerto Real, Haiti (Trevisan [1504] in
Symcox 2002:90). Many other references pertain
to the Lesser Antilles, but are worth including here
for comparative purposes. Breton (1997:9) noted
a dye being used to color “willow bark” which
“they painstakingly and with the greatest skill
divide... into extremely narrow bands... then soak...
in... red and black [dye], so that they produce a vari-
ety of designs..., [weaving] the strips like fabric.”
Cotton was also dyed: Rochefort (1666:258) notes
that men wore necklaces of shell, animal and pos-
sibly human bone “fasten’d together with a thread
of fine Cotton of a red or violet colour.” It is likely
that the practice of dyeing cotton and other fibers
was also prevalent among the Taíno to some degree,
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 311
Figure 8. A damaged area at the right shoulder reveals three layers of cotton: the inner-most layer (a), featuring a thick
cluster of single, Z-spun strands, is wrapped by evenly spaced two-ply, Z-spun, S-twist yarns (on the vertical) (b), which
serve as a warp for the more elaborate “embroidered” weft (on the horizontal) (c), creating the cemí’s “skin” (See
Supplemental Figure 8 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

312 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 9. Vegetable fiber cords used to bind the mandible to the embroidered “skin” and secure the teeth in place (note
that modern monofilament is also present from a recent conservation treatment). (See Supplemental Figure 9 for color
image.)
Figure 10. The lower left eye socket, with fiber cords just visible in the inner recesses that function to secure the outer
embroidery (see arrows). There is a notable gap between the black shell “pupil” and the outer eye frame, suggesting that
another shell layer representing the white sclera may have been present. (See Supplemental Figure 10 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

although there appears to be a preference for nat-
ural cotton finishes, at least among the elites: in His-
paniola cacique Behecchio’s thirty wives wore
white, elaborately woven naguas, while in Cuba,
a group of elite men reportedly wore long, white
tunics (Bernaldez 1967:142; Colón 1992:139; Las
Casas 1951:I:441; Angelo Trevisan [1504] in Sym-
cox 2002:93).
Once processed, hair or other materials could
have been combined with the cotton and beaten
together to make the fibers soft and pliable (cf.
Mastache 2005:86). Spinning the material into yarn
was a critical stage, as quality and strength of the
weave are based on the characteristics of the thread
(Mastache 2005:86). Spindle whorls were used in
this process; many ceramic, and to a lesser degree
shell, examples have been found in archaeological
sites dating from the late Saladoid period (ca. A.D.
500)—the former often small, biconical, and dec-
orated with red pigments and/or incised designs
(Boomert 2000:300; Rouse 1992:129). The size of
spindle regulated the thickness of the spun fiber—
the smaller the spindle, the finer the yarn. Spinning
gave the yarn a characteristic twist (either S or Z-
spun): in the case of the cotton cemí many of the
observed single strand yarns are Z-spun. Single
strands are predominantly seen in the padded knee
and arm wrappings, where the more decoratively
woven surfaces have been lost (Figure 8, Supple-
mental Figure 8). These range in size from .50 to
2 mm in thickness— suggesting either the use of a
variety of spindle sizes, or natural variations within
the relatively quick spinning of single strands.
These yarns needed to be made in bulk, for not only
were they laid down in quantity in areas below the
more decorative weaving, but they had to be plied
to make stronger yarns that made up the majority
of surface weave, including the textured “skin” of
the cemí, which is constructed of two ply Z-spun,
S-twist yarns.
Although cotton made up the cemí’s outer sur-
face, a variety of materials went into its internal
construction. CT scans of the cemí reveal that veg-
etable fiber cords were used to tightly bind the
inner structure together, criss-crossing between
the bound canes forming the framework of the legs
and arms, creating a solid foundation for the cot-
ton overweave. Fiber cords were also used to
secure the woven “skin” to the human skull— as
is seen in the binding around the mandible and
maxilla and within the eye sockets (Figures 9, 10,
and 19 and Supplemental Figures 9, 10, and 19).
The fiber cords, which are 2 ply Z-spun and quite
fine in thickness (ca. .50 mm), have yet to be iden-
tified botanically: although a variety of materials
were likely used for cordage, the cronistas do note
the use of cabuya (Oviedo 1992:I:117; Furcraea
sp.? [Moscoso in Tejera 1977:243] or other plants
in the agave family) and “linen-like” grasses that
the natives “braid [into] ropes much stronger and
tauter than hemp” (Scillacio [1494] in Symcox
2002:45). The strength of the cords was clearly
critical in some areas: the mouth was deliberately
woven open to reveal the teeth, each of which has
been secured with cordage (Figure 9, Supple-
mental Figure 9), suggesting that the complete-
ness and durability of this area was crucial to the
final appearance of the cemí, conforming to the
cannons of Taíno style, which emphasized the
mouth and teeth.
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 313
2 5 4
3
stage 1
Figure 11. Suggested sequence of construction.
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

The Manufacture Process
Once the various materials had been processed in
sufficient quantities, the construction could begin
(Figure 11). It is unclear what protocols regulated
the creation of these figures— whether the work
was done in isolation, through ritually sanctioned
ways, and/or only by certain individuals (e.g.,
behiques). However, the skill in bringing these
materials together in a construction that withstood
centuries, despite being made almost entirely of
delicate perishables, would suggest the work of an
experienced artisan, well versed both in weaving
techniques and sculptural composition. The con-
struction was undoubtedly planned in advance. This
is most readily seen in the positioning of “weights”
placed within the cemí’s body cavity, which CT
scans reveal to be two different materials—a large
stone positioned at the front (belly and chest) and a
paddle-shaped piece of wood placed at the back, its
narrowing tip extending into the neck and head (Fig-
ure 4, Supplemental Figure 4) (Martina et al. 2010:
Figures 4 and 5; Vega 1987). These two elements
provide balance and support when the cemí is in an
upright, seated position: the posture appears per-
fectly centered when the haunches are raised slightly
on a support— this better aligns the body, bringing
the head forward and the body upright (in contrast,
see Kerchache 1994:160–161, where the cemí is
positioned on a flat surface, resulting in a precari-
ous, backwards lean). The posture— with bent knees
and straight back— combined with the specific ori-
entation of the stone weight suggest that the cemí’s
upright sitting position was likely intentional, with
another object, such as a ceremonial seat or duho,
supporting it when on display or in “use.”
Around this foundation of stone and wood, the
armature of the sculpture was built: bundles of long
lianas or canes were bound to the sides, becoming
the basis of the arms and legs (Figure 11.1). These
multiple strands ran the length of the appendages,
reinforcing each other but also providing some
flexibility. Some canes were separated at the hands
and feet to become the foundation for the fingers
and toes (Figure 12, Supplemental Figure 12), and
CT scans reveal the presence of wooden frame-
works at the feet, to provide the flattened shape of
the soles (Martina et al. 2010:Figure 7). At various
314 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 12. Internal cane supports for the fingers are visible at the tips of the second and fifth digits of the left hand. (See
Supplemental Figure 12 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 315
Figure 13. Two ply Z-spun, S-twist yarns in decorative checked-weaves at the left knee. In the upper left are remnants of
a black resinous material that adheres to much of the inner knee areas. (See Supplemental Figure 13 for color image.)
Figure 14. A damaged area at one of the right ribs reveals an internal cane used to create the skeletal imagery on the sur-
face of the cemí (e.g., ribs, hips). (See Supplemental Figure 14 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

points along the length of the body, thick vegetable
fiber cords bind these elements tightly, creating a
solid, internal framework. The body and limbs were
then padded with either unprocessed or loosely
spun cotton (Figure 11.2), with a heavy single-ply
cotton twine, securing the various elements into
place as well as serving as the warp for the “embroi-
dered” weave of the cemí’s “skin” (Figure 8, Sup-
plemental Figure 8). Cotton strands serving as the
weft were then added and the weave tightened con-
siderably (Figure 11.3). It is unclear as yet what
tools may have been used in this process, although
needles (Allsworth-Jones 2008:182) and awls are
known from the archaeological record. As con-
struction progressed, the weave expanded or con-
tracted around the body’s curves— analogous to a
sprang weaving technique (Collingwood 1974). In
keeping with the Taíno tradition of tightly binding
the upper arms and knees,3 these areas were kept
quite narrow on the cemí’s body, woven with only
minimal internal padding, in contrast to the more
“fleshy” areas. These constricted areas were then
enhanced with additional layers of spun cotton, the
under layer consisting of a roughly spun, single-
ply cotton, over which were added stronger two-
ply yarns in a diamond pattern (Figure 13,
Supplemental Figure 13, Figure 11.4). Among the
finishing touches, flexible canes were added to the
surface to delineate eye sockets, lips, ears, ribs, hip
joints, navel and chest, some of which emphasize
key skeletal elements that feature strongly in Taíno
iconography (Figure 14, Supplemental Figure 14,
Figure 11.5).
The frayed seams midway down the back of the
neck (Figure 15, Supplemental Figure 15) suggest
that the head was most likely woven separately, and
added to the body at the final stages of construc-
tion. This makes sense given the level of access
required to weave the skull in the round. A series
of thick cotton cords, now broken, extended from
the back of the jaw around the neck to fasten the
head more securely to the neck. The reliquary’s
inner structure can be seen beneath the large woven
band at the back of the head, including the parietal
316 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 15. Back view of the cemi, where a frayed end seam overlaps the neck, suggesting that the head was woven sepa-
rately. The strap emerges from between the shoulder blades, its center currently attached to the top of the head, perhaps
in efforts to cover and secure the damaged area, which exposes a group of gourd panels. (See Supplemental Figure 15
for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

bones and the presence of what appear to be thin
panels of gourd (possibly calabash, Crescentia
cujete) (Figure 15, Supplemental Figure 15). Much
of the back of the skull had been removed, appar-
ently being too large and/or deemed unnecessary
for the purposes of the sculpture. The inclusion of
the gourd fragments may have functioned to help
maintain a rounded appearance to the back of the
head while perhaps also acting as a symbolic ref-
erent to the more common practice of placing skulls
in gourds (Colón 1992:75, and 114–116).
The woven band currently attached to the top of
the head was originally a loose strap that fell behind
the cemí, possibly to secure or suspend it, with its
terminal ends secured between the shoulder blades
(early images— such as Cronau 1892:I:263; Fewkes
1891:Figure 7; and Seelye 1892:133—feature the
top of the head free of the band). Its current place-
ment, attached to the top of the head by modern
thread, appears to be a recent conservation treatment
to safeguard the damaged area exposing the skull
and loose gourds. This damage is old, perhaps a
result of early attempts (pre-1891; see Fewkes
1891:Figure 7) to see what was inside the figure,
and the use of the strap to cover this area has a long
history. It first appears low on the forehead in the
British Museum’s archival photograph (Figure 3),
and over the years appears to migrate to different
positions on the forehead (e.g., see Vega 1987:Fig-
ure A-B, who calls it a “turbante”).
The entire object may have then been selec-
tively stained to enhance certain features— as there
are clearly brighter and darker areas on the surface
(Figure 11.5; Ripley 1980:117; Vega 1987:16). For
example, some of the multi-strand cotton “bun-
dles” added to the sides of the leg and shoulder
bands appear stained dark at the front (possibly
with resins), and white at the back (Figure 16, Sup-
plemental Figure 16). There are other color varia-
tions visible— from white to a brown- yellow—
although it is not clear at this stage whether some
are a result of depositional damage (e.g., bat urine
and fungal decay can bleach cotton), intentional
staining, or indeed, a combination of the two. A yel-
low ochre stain appears on raised bands at the navel,
chest, forehead and ears, possibly a result of spe-
cific staining to these areas. Naturally occurring
colorants such as red ochre (used in Caribbean bur-
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 317
Figure 16. Bundle of cut cotton twine at the right shoulder, covered with black pigment or resin at the front, with white
material covering the back (length: 22.45 mm; width: 18.56 mm). (See Supplemental Figure 16 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

ial contexts since the Archaic— e.g., Schaffer et al.
2010) and the seeds of Bixa orellena, mentioned
above, yield a vibrant red pigment that can tinge
yellowish-brown over time (Yde 1965:64). Future
analysis will hopefully shed light on this issue.
The cemí features areas covered with a thick,
dark substance, notably around the inner legs, navel,
palms, and, to a lesser degree, on certain parts of
the upper body, although it is not clear as yet whether
these are all the same material. A thick layer within
the navel and palms suggests that it may have func-
tioned to adhere inlays (Figure 17, Supplemental
Figure 17). A small amount of thisdark material (1.2
mg) was noted on the cotton sample from the knee
and was separated for analysis, revealing a lipid
composition together with Protium or Bursera
genus triterpenoid resin (Erika Ribechini, personal
communication 2010), adhesives also found in other
Taíno sculptures (Ostapkowicz et al. 2012:2245).
In the Caribbean, Bursera resin has been used for
a wide variety of purposes, ranging from medicine
to glue (Gibney and White in Nicholls 2006:17;
Little and Wadsworth 1964:236; Timyan 1996:210).
There appear to be other, lighter colored sub-
stances—perhaps also resinous—lodged in the
deeper crevices of the sculpture, although these still
require identification.
Resins are also present in the eyes and mouth.
The right eye consists of a large perforated white
shell held in place by amber-like resin granules
(Figure 18, Supplemental Figure 18), a treatment
that appears in striking contrast to the left eye, made
of a large black unperforated shell lodged within
dark resins. There has been much speculation
regarding the meaning behind such a juxtaposition
of light and dark materials (e.g., see Royo Guardia
1947:150, who suggests it was symbolic of the
cemí’s diurnal/nocturnal vision). But there are
grounds to suggest that there has been some dam-
age to the left eye, in the form of a gap between the
woven framework and the black shell, as if another
layer may have been present (see Figure 10, Sup-
plemental Figure 10). There are also small resinous
granules adhering to the black shell’s surface, but
absent from its raised, pupil-like center. Together,
this suggests that there may have been a layering
of materials in the eye, with a soft resin first placed
in the cavity to infill the area and to adhere the dark
318 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 17. Thick black substance placed within the palm of both hands and the navel, suggesting that these areas may
have held inlay. (See Supplemental Figure 17 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

shell, to which a drilled white shell was attached
with further resin, so that each eye would have a
dark pupil and white sclera. On the other hand,
there is no such black shell in the right eye, if indeed
it was ever present. Thus, the original appearance
of the eyes remains unclear. A dark resin-like coat-
ing also covers the upper palate (Figure 19, Sup-
plemental Figure 19), suggesting that the inner
mouth was intended to recede from view, enhanc-
ing the whiteness of the teeth.
The above additions to the cemí’s surface—
pigments, resins, shell— are all that remains of what
was originally an object adorned with materials of
contrasting colors and textures against its freshly
woven cotton “skin.” The ear lobes, for example,
were woven as loops (now broken) to hold ear-
spools of carved wood, stone, shell, or feather bun-
dles. A large shell or perhaps guanin (a gold-copper
alloy) may have been featured in the navel, while
the unusual placement of resins in the palms sug-
gests that these areas were also highlighted with
inlays, marking them as visually and symbolically
important. Both the navel and hands were focal
areas within Taíno sculpture: raised palms reoccur
in ceremonial objects, including cohoba stands
(e.g., wooden sculpture in Figure 3) and vomiting
spatulas, and their position likely had a ritual sig-
nificance. The navel is an important feature in
anthropomorphic carvings, often enlarged for
effect. According to Pané (1999:18–19) it indicated
that the individual was alive, in contrast to the dead
(operito), who had no navels. Its presence on a
sculpture encasing the bones of an ancestor may
have served to underline their still vital involve-
ment with the world of the living.
Discussion: Creating an Ancestor
This sequence of production— from gathering cot-
ton to adding the last inlay— is seemingly straight-
forward; but this is no simple object. The materials
combine to create a remarkable “being”—a phys-
ical representation of an ancestor. The cotton weave
becomes a metaphor for his skin; the stone weight,
his viscera; the lianas, the bones and sinews of his
arms and legs; the wood, his backbone. The human
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 319
Figure 18. The cemí’s right eye (seen here on the left) features granules of amber resin (likely a result of deterioration),
covered by a white perforated shell, while the black shell in the left eye (seen here on the right) is held in place by dark
resins, its outer surface also covered by resinous adhesions, suggest that another layer—now lost—may have been added.
(See Supplemental Figure 18 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

form is reconstituted, made complete again in the
hands of the weaver who him/herself becomes dur-
ing this process a “creator” of sorts, with the power
to bring together inanimate matter that becomes
embodied and animated by an ancestral force. If
the weaver was female— and given the cronistas’
accounts of women weaving, this is possible (Las
Casas in Báez Díaz 1977:12; Oviedo 1992:I:117;
for Lesser Antilles see Rochefort 1666:295; for
general discussion see Sued-Badilllo 1993:36–39;
Ostapkowicz 1998:527–533)—this further builds
on the reproductive symbolism of this act: of a
symbolic gestation and rebirth into the ancestral
realm. The physical process of creation, of build-
ing and adorning the body, must have had signifi-
cance now difficult to assess. This appears, at least
superficially, to be different from the carving of a
wooden cemí, which is a subtractive process that
is understood to release— or “realize”—the numi-
nous force inherent in the material (Pané
1999:25–26). In contrast, creating a cotton reli-
quary was an “additive” task, essentially building
on a foundation of bone and other materials— to
bind the forces within. It may have involved a
deeper significance given that the work was
based— literarily and metaphorically— on the
physical remains of a known individual. The
remains were worked into something that tran-
scended the familiar to become a direct link to the
numinous. When and how did this transition take
place? Were ceremonies performed to “invest” the
cemí— as they were for wooden cemís (Pané
1999:25–26), or was this numinous essence already
inherent in the remains? At what point in the
processes does the cemí become “cemífied’? We
can now only tentatively glimpse the undoubtedly
rich meanings behind such a creation— and engage
with the nature of what it may have meant to “cre-
ate,” or bring into physical being, an ancestor.
At a basic level, there are numerous references in
the early cronistas to suggest that wrapping and
binding with textiles was culturally significant, sep-
arating and protecting the individual or object.
Pané, for example, notes that stone cemís were
wrapped in cotton: “they take good care of them,
wrapping them in cotton, placing them in small bas-
kets, and putting food before them; they do the
same with the cemís they have in their houses” (in
Colón 1992:160; see Walker 1993:47–48). Larger
cemís would have been wrapped in various ways—
320 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Figure 19. Interior view of the mouth showing black resinous material covering the upper palate and fiber cords hold-
ing the teeth in place. (See Supplemental Figure 19 for color image.)
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

including cotton ornaments suspended from their
perforated ears, and wrapped around their arms and
legs. Other substances, such as guanin, were also
wrapped in cotton— as in the case of a large disc
covered with four layers of cloth excavated from
the site of El Chorro de Maíta, Cuba (Guarch Del-
monte in Oliver 2000:201). Wrapping these impor-
tant cemís/objects in protective material functioned
to separate them, marking them as precious. But it
was not just objects that were covered in cotton:
Oviedo (1992:I:119) mentions a burial involving
very long cotton bandages that were wrapped
tightly around the deceased prior to his interment.
There is ample archaeological evidence for the
practice of wrapping human remains, spanning
much of the Caribbean region— from the Saladoid
site of Punta Candelero, Puerto Rico, where 80 per-
cent of burials were reportedly bound with veg-
etable fibers or hammocks and interred in squatting
positions (Rodríguez 1997:82) to Taíno satellite
sites such as Kelby’s Ridge, Saba, where crema-
tions were suggested to have been interred wrapped
in cloth (Hoogland and Hofman 1993:170). There
may have been a range of practices, from elite buri-
als featuring a wealth of cotton goods, to more
modest burials for commoners, where individuals
were accompanied by only their cotton arm or leg
bands. Thus, cotton appears much more than a com-
modity, adorning both the living and the dead. It
may well have been, in specific contexts, a trans-
formative material, marking life transitions. Gifts
of cotton cloth may have been given during the
most important stages of life: birth, initiation and
marriage, as well as forming the final parting gift
to the dead (cf. Murra 1962).
And who was the individual bound in the cot-
ton “skin” of the Turin cemí? The skull is proba-
bly that of an adult male, based on the morphology
of the supraorbital ridge and the mandible (Mar-
tina et al. 2010:1994). The teeth are minimally
worn with the only visible spots of dentine present
on the first mandibular molar; the third molars are
missing congenitally, as shown in a CT scan (Mar-
tina et al. 2010:Figure 2a). While two maxillary
teeth, the first left molar and the second right pre-
molar, were lost ante-mortem, the overall impres-
sion is that of a young adult (Rick Schulting,
personal communication 2007), particularly as pre-
colonial Caribbean populations experienced fairly
heavy dental wear due to sand and other abrasives
present in the diet (Crespo-Torres 2010:202). The
skull exhibits evidence for cranial modification,
which would have been done during the individ-
ual’s infancy (Martina et al. 2010:1995, 1998). Cra-
nial modification among the Taíno appears to have
been a fairly common practice, and may have been
a marker of group identity rather than social status
(Crespo-Torres 2010:206), although these distinc-
tions are still poorly understood.
The fact that this individual’s skull had been
incorporated into so elaborate a cemí suggests that
he had attained some standing in the community.
Caciques were thought to have been the main recip-
ients of such treatment (Las Casas 1997:176; Loven
1979:586; Vega 1987:13). As the Geraldini quote
that opened this paper makes clear, only those that
were “religious and just and ... conferred some ben-
efit on the people” were so honored. In contrast,
there was a wide diversity of burial styles in the
Caribbean, including secondary internments where
various skeletal elements were re-buried. Burials
were opened, and skeletal elements— often the
skull, but sometimes a femur or other long bones—
were removed; this practice is known archaeolog-
ically across much of the Caribbean, from Trinidad
and Tobago (Boomert 2000:398–400) to Eleuthera,
Bahamas (Schaffer et al. 2010), and mentioned
with some frequency in cronistas’ accounts from
Cuba to Guadeloupe (Colón 1992:75, 114–116;
Las Casas 1967:I:246). In some instances skull
fragments and long bones were carved with images
and made into pectorals or ornaments, although it
is still unclear whether this treatment was reserved
for the remains of enemies or ancestors (Roe 1989).
At the site of Paso del Indio, Puerto Rico, where
138 interments were found, many primary burials
were directly associated with secondary burials
(Walker 2005:73). Three of these particularly stand
out: each contains a primary burial of an adult male
holding a secondary skull of another male in his
arms. The intimate nature of their position— with
the skulls cradled closely against their bodies—
suggests to Walker (2005:74) that they were “trea-
sured and dear”: perhaps ancestral remains, rather
than trophies from slain adversaries (see also Oliver
2009:146–147).
The cotton cemí offers a rare window onto
another funerary practice: the elaborate nature of
the figure implies that it was made to be visible—
whether to a restricted audience or the whole
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 321
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

322 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
community— and given its sturdy construction, it
was meant to have a long “life.” The work involved
in its creation went far beyond the wrapping of
bones in a piece of cotton cloth— manifesting
instead an entire body that could sit upright, with
limbs that were, to some limited degree, movable.
The ancestor was thus “reconstituted”—created
anew through layers of materials. This tangible
form kept the history of a specific individual’s
deeds alive in the minds of the living, reinforced
through oral accounts passed down generations.
Through their previous actions and their transcen-
dance into a liminal state— of the spirit world but
also reconstituted in an earthly guise— these ven-
erable beings had amassed the knowledge to guide
actions. Ethnographic accounts make clear that
these figures were oracles, offering advice and
counsel— perhaps in this sense we can begin to
understand why the figure’s mouth was paid such
special attention, and was woven open, as if speak-
ing. And it is this effort of harnessing the
numinous— of seeking assistance and guidance or
tapping into the potent powers of those whose deeds
were legendary— that underscores the cemís’ pur-
pose. They were viewed as active agents with the
power to affect events— agents that undoubtedly
needed to be appeased in efforts to assure their
goodwill. Such forces were treated with great
respect, their names and actions recounted, their
presence central to certain ceremonies, likely sym-
bolically “fed” and otherwise taken care of— to
maintain the beneficial flow of things. This in
essence made them central to events and involved
in the lives of their descendants and community
members. Their “death” may simply have been
viewed “as another rite of passage into the next
stage of the life/death/rebirth cycle. Such ancestors
remain[ed] relevant to and concerned with, or pro-
tective of, the living among whom they [were]
passed down as inheritable relics” (Roe 1989:859).
Using such intermediaries appears a universal prac-
tice, both in traditional cultures and inherent in
many of today’s major religions. As Holsbeke
(1996:15) points out:
By giving intangible beings a material shape,
their supernatural power is localized, the bet-
ter to control it. At the same time the believer
can tap the spiritual energy of the statue, which
brings him into direct visual and tactile con-
tact with the being represented. It functions as
a bridge to the supernatural, and as a window
through which the believer can look at the
other world, while at the same time being seen
from the other side.
And the critical connection between the ancestral
cemís and descendants made them unique among
the multitude of other Taíno cemís. According to
Pané (1999), other cemís were active agents, ani-
mating wood and stone at whim, apparently with-
out any direct connection to their eventual
custodian. They were capricious forces, some
choosing to escape servitude to those who could
not appease or placate them (Pané 1999:25–26). In
contrast, reliquaries and cotton cemís were con-
nected to a place and community via their descen-
dants. These not only linked with the living, but
also to other deceased relatives that had been cemíi-
fied, all connected “in a web of kinship and
descent”—as opposed to a trustee relationship that
existed with other forms of cemí “ownership”
(Oliver 2009:62). As Oliver (2009:141) notes, the
cotton cemí, “as a cemíified ancestor, defines the
relationships (duties, obligations, modes of social
conduct) among the surviving, living descendants.
. . . [and] has the potency to affect the production
and reproduction of the descendants, and the wider
community.” With such intimate links between the
ancestors and the living, these cemís, unlike oth-
ers, would not circulate in exchanges, but likely
remained within the descent group (Oliver
2009:250).
As carefully curated and safeguarded lineage
“treasures,” accruing greater potency with each
beneficial “act,” such cemís would be expected to
have a long life of involvement and ceremonial
activity. The older they became, the greater their
potency, and by association, the greater the pres-
tige of the descendants. It is quite possible that the
belief in their efficacy may have spanned genera-
tions. When not brought out for rituals and con-
sultation, which likely included the use of cohoba
(the ingestion of hallucinogenic drugs), they would
have been safeguarded in a cacical caney or other
secure locations, such as caves. As such, their his-
tories and reputations could have been built over
centuries. The date of A.D. 1439–1522 for the Turin
cemí, however, would suggest that its “afterlife”
was cut short by the arrival of the Spanish.
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 323
Summary and Conclusions
The cotton cemí was an investment on many lev-
els: of time, effort, artistry, knowledge, and belief.
It embodied these aspects through a fusion of the
now indefinable (ritual, ceremony) with the tangi-
ble: the layering of materials, from wood, stone,
lianas, vegetable fibers, cotton in quantity, animal
hair, gourd, resins, inlays and pigments, and, at its
core, the human skull. The creative vision and skills
of the artist/weaver are evident in each phase of
construction— from the foundation balanced by
weights and supports to the rich finishes (texture,
color, inlays) covering the surface. The elaboration
of the woven body, juxtaposing a life-affirming
physicality (fleshed body, enlarged navel, genitals)
with both skull and skeletal imagery (rib cage, hip
bones), functioned to bring the forces of life and
death clearly and eloquently into view. This cre-
ative and technical investment ultimately reflects
the cemí’s deep social and cultural value: one based
on the belief of an unbroken link between the liv-
ing and the dead. The integration of ancestral
remains and relics into both ritual and the every-
day functioned on a variety of levels: from linking
descendants with powerful predecessors and so
establishing and reinforcing status and ancestral
legacies, to harnessing these ancestral/cemí forces
as intermediaries for the benefit of the community.
In this sense and many others, this figure was—
and remains— a bridge between the living and those
who have gone before.
Acknowledgments. The study of the Turin cemí was made
possible through the support of the Leverhulme Trust and the
Getty Foundation, with NERC/ORADS supporting the
radiocarbon dating. We are grateful for the generous assis-
tance of Prof. Emma Rabino Massa, Director, and Dr. Rosa
Boano, Assistant Director, MAET, during Ostapkowicz’s
visits to the collections in 2005 and 2007. Our thanks to Rick
Schulting for his observations on the visible skeletal
remains, and comments on previous versions of this paper;
Candace McCaffery, Florida Museum of Natural History, for
facilitating loans of animal hair samples from the museum’s
collections; Fiona Brock and Tom Higham from the Oxford
Radiocarbon Laboratory for Archaeology and the History of
Art; Erika Ribechini, University of Pisa, for the GC/MS
resin analysis results; Sarah Mallet for assistance with the
translation of du Tertre’s text on vengeance killings (other
translations by Ostapkowicz); Francisca Santana Sagredo for
translation of the abstract into Spanish; Donatella Minaldi
and Raphfaella Bianucci for their previous assistance; Dr.
María Cristina Martina for permissions to reproduce Figure
4 and for discussions on CT images of the infraorbital fora-
men; Colin McEwan for permission to use Figure 3 and Dan
Bruce and Lynda Miller for their assistance with, and
insights of, this archival photograph; José Oliver and two
anonymous reviewers for their comments. Unless otherwise
noted, all photographs and illustrations are by Ostapkowicz,
and are courtesy of the Museum of Anthropology and
Ethnography, Turin.
Supplemental Materials. Supplemental materials are linked to
the online version of the paper, which is accessible via the
SAA member logon at http://saa.org/Home/tabid/36/
Default.aspx.
References Cited
Alegría, Ricardo E.
1980 Etnografia taína y los conquistadores. In Proceedings
of the 8th International Congress for the Study of Pre-
Columbian Cultures of the Lesser Antilles, edited by
Suzanne M. Lewenstein, pp. 430–446. Anthropological
Researc h Papers No. 22. Arizona State University,
Phoenix.
1981 El uso de la incrustación en la escultura de los indios
antillanos. Centro de estudios avanzados de Puerto Rico
y el Caribe con Fundación García Arévalo. I.G. Manuel
Pareja, Barcelona.
Allsworth-Jones, Phillip
2008 Pre-Columbian Jamaica. Caribbean Archaeology and
Ethnohistory, L. Antonio Curet, series editor. The Uni-
versity of Alabama Press, Tuscaloosa.
AMT (Amici del Museo di Torino)
ca. 1975 Gli Zemi delle Antille: 4B, Museo di antropologia
ed etnografia, Torino. Libreria Editrice Universitaria Lev-
rotto & Bella, Torino.
Báez Díaz, Tomás
1977 La mujer aborigen y la mujer en la Colonia: su par-
ticipación en la etapa histórica que termina en el 1795.
Firma del Tratado de Bas ilea, Premio Shell, Sant o
Domingo.
Bernáldez, Andrés
1967 History of the Catholic Sovereigns Don Ferdinand and
Doña Isabella. In Select Documents Illustrating the Four
Voyages of Columbus, edited by Cecil Jane, pp. 114–168.
2nd ed. Hakluyt Society, Kraus Reprint, Millwood, New
York.
Biscione, Marco
1997 Epilogue: The Beaded Zemi in the Pigorini Museum:
History and Problems of Interpretation. In Taíno: Pre-
Columbian Art and Culture from the Caribbean, edited by
Fatima Bercht, Estrellita Brodsky, John Alan Farmer, and
Dicey Taylor, pp. 158–163. The Monacelli Press, New York.
Boomert, Arie
2000 Trinidad, Tobago and the Lower Orinoco Interaction
Sphere: An Archaeological/Ethnohistorical Study. Cairi
Publications, Alkmaar.
Breton, Rev. Fr. Adrien Le
1997 His tor ic Acc oun t of Saint Vinc ent the India n
Youroumaÿn, the Island of the Karaÿabes, by Rev. Fr.
Adrien Le Breton S.J. (1662–1736). The Mayreau Envi-
ronmental Development Organization, Kingstown, St. Vin-
cent and the Grenadines.
Collingwood, Peter
1974 The Techniques of Sprang: Plaiting on Stretched
Threads. Faber and Faber, London.
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

324 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012
Colón, Fernando
1992 [1538?] The Life of the Admiral Christopher Colum-
bus by his Son Ferdinand. Translated by Benjamin Keen
[1959]. Rutgers University Press, New Brunswick, New
Jersey.
Crespo-Torres, Edwin F.
2010 Ancient Bones Tell Stories: Osteobiography of Human
Remains from Tibes. In Tibes: People, Power, and Ritual
at the Center of the Cosmos, edited by Antonio Curet and
Lisa M. Stringer, pp. 191–208. The University of Alabama
Press, Tuscaloosa.
Cronau, Rudolf
1892 Amerika. Die Geschichete seiner Entdeckung von der
altesten bis auf die neueste Zeit. Abel and Mueller, Leipzig.
Curet, Antonio, and José R. Oliver
1998 Mortuary Practices, Social Development, and Ideol-
ogy in Precolumbian Puerto Rico. Latin American Antiq-
uity 9:217–239.
du Tertre, R. P., Jean- Baptiste
1667 Histoire Générale des Antilles Habitées par les
François, Tome II: Conte nant L’Histoire Naturelle.
Thomas Lolly, Paris.
Feest, Christian
1986 Zemes Idolum Diabolicum. Archiv für Völkerkunde,
40:181–198.
1991 Beaded Belt. In Circa 1492: Art in the Age of Explo-
ration, edited by Jay A. Levenson, pp. 580–581. National
Gallery of Art, Washington, D.C.
Fewkes, Jesse Walter
1891 On Zemes from Santo Domingo. American Anthro-
pologist 4:167–176.
1903 ms. 4408(45a) Diary of Archaeological notes and
drawings, Nov. 15–Jan. 16/03, 1 vol, ca. 155pgs, Puerto
Rico and Santo Domingo. Ms. on file at the National
Anthropology Archives, Smithsonian Institution, Wash-
ington, D.C.
1907 The Aborigines of Porto Rico and Neighbouring
Islands. In 25th Annual Report of the Bureau of American
Ethnology, 1903-04. Smithsonian Institution, Washing-
ton, D.C.
García Arévalo, Manuel A.
1997 The Bat and the Owl: Nocturnal Images of Death. In
Taíno: Pre -Columbian Art a nd Culture from th e
Caribbean, edited by Fatima Brecht, Estrellita Brodsky,
John Alan Farmer, and Dicey Taylor, pp. 112–23. The
Monacelli Press and El Museo del Barrio, New York.
Girotti, Marilena, and Donatella Meaglia
2001 Zemi in cotone del Museo di Antropologia di Torino:
Storia del ritrovamento e descrizione. In Zemi a Torino.
Museo di Antropologia ed Etnografia dell’Università di
Torino, Turin.
Guidi, Annalena, and M. Grazia Appendino
1973 I Taíno ed I loro zemi. Rivista di Etnografia,
XXVII:3–11.
Hedges, S. Blair
2001 Caribbean Biogeography: An Overview. In Biogeog-
raphy of the West Indies: Patterns and Perspectives, edited
by Charles. A. Woods and Florence. E. Sergile, pp. 15–34.
CRC Press, Boca Raton, Florida.
Holsbeke, Mireille (editor)
1996 The Object as Mediator: On Transcendental Mean-
ing of Art in Traditional Cultures. Etnografisch Museum
Antwerp, Antwerp.
Hoogland, Menno L. P., and Corinne L. Hofman
1993 A 14th-Century Taíno Settlement on Saba, Netherlands
Antilles. Analecta Praehistorica Leidensia 26:163–181.
Kerchache, Jacques (editor)
1994 L’Art Taïno: L’Art des Sculpteurs Taïnos Chefs-
D’œvre des Grandes Antilles Precolombiennes. Musees de
la Ville de Prais, Musée du Petit Palais, Paris.
Langer, Reinhart Hug Michael, and George D. Hill
1991 Agricultural Plants. 2nded. Cambridge University
Press, Cambridge.
Las Casas, Fray Bartolomé de
1951 Historia de las Indias. Biblioteca Americana, Cro-
nis tas de Indi as series, 3 vols. Fondo de Cultura
Económica, Mexico City
1967 Apologética historia sumaria. Universidad Nacional
Autónoma de México, Instituto de Investigaciones Históri-
cas, México, D.F.
1997 Chapter 120: Concerning the idols worshipped by the
Indi ans of the Island of Hisp aniola. In Taíno: Pre-
Columbian Art and Culture from the Caribbean, edited by
Fatima Brecht, Estrellita Brodsky, John Alan Farmer and
Dicey Taylor, pp. 112–23. Translated by Susan C. Griswold.
The Monacelli Press and El Museo del Barrio, New York.
Little, Elbert L., and Frank H. Wadsworth
1964 Common Trees of Puerto Rico and the Virgin Islands.
Agricultural Handbook No. 249. U.S. Department of Agri-
culture, Washington.
Loven, Sven
1979 Origins of Tainan Culture, West Indies. 2nd ed. AMS
Press, New York.
Martina, María Cristina, Federico Cesarani, Rosa Boano, Emma
Rabino Massa, Claudio Carlo Venturi, and Giovanni
Gandini
2010 Scenes from the Past: Multidetector CT of an Antil-
lean Zemi Idol. Radiographics 30:1993–1999.
Masali, Melchiorre, and Gabriella Erica Pia
1991 I materiali taíno del Museo di Antropologia di Torino:
lo Zemi di cotone. In Gli Indios di Hispaniolae la prima
Col óni zza zione euro pea . L’Un ive rso, supp. al No.
1/1991:85–90.
Mastache, Guadalupe
2005 Weaving in Ancient Mexico. Arqueología Mexicana.
Edición Especial, Vol 19.
Martyr D’Anghera, Peter
1970 [1530] De Orbe Novo: The Eight Decades of Peter
Martyr D’Anghera. Translated by Francis Augustus Mac-
Nutt. 2 vols. Burt Franklin, New York.
Meaglia, Donatella, and Marilena Girotti
2001 Zemi a Torino. Museo di Antropologia ed Etnografia
dell’Università di Torino, Turin.
Murra, John V.
1962 Cloth and Its Functions in the Inca State. American
Anthropologist 64:710–728.
Newsom, Lee Ann
2008 Caribbean Paleoethnobotany: Present Status and New
Horizons (Understanding the Evolution of an Indigenous
Ethnobotany). In Crossing the Borders: New Methods and
Techniques in the Stu dy of Mate rial Cul ture in the
Caribbean, edited by Corrine L. Hofman, pp. 173–194.
The University of Alabama Press, Tuscaloosa.
Newsom, Lee Ann, and Elisabeth S. Wing
2004 On Land and Sea: Native American Uses of Biotic
Resources in the West Indies. University of Alabama Press,
Tuscaloosa.
Nicholls, Robert W.
2006 Remarkable Big Trees in the U.S. Virgin Islands. Uni-
versity of the Virgin Islands and the Urban and Commu-
nity Forestry Assistance Program, VI. Department of
Agriculture, St Thomas.
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

Olazagasti, Ignacio
1997 The Material Culture of the Taíno Indians. In The
Indigenous People of the Caribbean, edited by Samuel Wil-
son, pp. 131–139. University Press of Florida, Gainesville.
Oliver, José R.
2000 Gold Symbolism among Caribbean Chiefdoms: Of
Feathers, Cibas and Guanín Power among Taíno Elites. In
Precolumbian Gold: Technology, Style and Iconography,
edited by Colin McEwan, pp. 196–219. British Museum
Press, London.
2009 Caciques and Cemí Idols: The Web Spun by Taíno
Rulers Between Hispaniola and Puerto Rico. The Uni-
versity of Alabama Press, Tuscaloosa.
Ostapkowicz, Joanna
1998 Taíno Wooden Sculpture: Duhos, Rulership and the
Visual Arts in the 12th–16th Century Caribbean, unpub-
lish ed PhD dissertation, University of East A ngli a,
Norwich.
Ostapkowicz, Joanna, Christopher Bronk Ramsey, Fiona Brock,
Tom Higham, Alex C. Wiedenhoeft, Erika Ribechini, Jean-
nette J. Lucejko, and Samuel Wilson
2012 Chronologies in Wood and Resin: AMS 14C Dating
of pre-Hispanic Caribbean Wood Sculpture. Journal of
Archaeological Science 39:2238–2251.
Ostapkowicz, Joanna, Alex Wiedenhoeft, Christopher Bronk
Ramsey, Erika Ribechini, Samuel Wilson, Fiona Brock,
and Tom Higham
2011 ‘Treasures... of Black Wood, Brilliantly Polished’:
Five Examples of Guaiacum Sculpture from the Tenth to
Sixteenth Century Caribbean. Antiquity 85(329):942–959.
Oviedo y Valdés, Gonzalo Fernández de
1992 [1535–49] Historia general y natural de las Indias.
5 vols. Biblioteca de Autores Españoles. Ediciones Atlas,
Madrid.
Pané, Fray Ramón
1999 An Account of the Antiquities of the Indians. Edited
by José Juan Arrom, and translated by Susan C. Griswold.
Duke University Press, London.
Petraco, Nicholas, and Thomas Kubic
2004 Color Atlas and Manual of Microscopy for Crimi-
nalists, Chemists, and Conservators. CRC Press, Boca
Raton,Florida.
Pignoria, Lorenzo
1626 Imagini de gli dei Indiani.In Seconda Novissima Edi-
tione delle Imagini de gli dei delli Antichi, edited by Ven-
cenzo Cartari, pp. 545–589. Nella stamparia di Pietro Paolo
Tozzi, Padoua.
Rehm, Sigmund, and Gustav Espig
1991 The Cultivated Plants of the Tropics and Subtropics:
Cultivation, Economic Value, Utilization. Verlag Josef
Margraf, Weikersheim, West Germany.
Ripley, Geo
1980 El cemí taíno de algodón. Boletín de Museo del Hom-
bre Dominicano 13:115–124.
Rochefort, Charles de
1666 The History of the Caribby-Islands, Viz. Barbados, St.
Christopher, St. Vincents, Martinico, Dominicano, Bar-
bouthos, Monserrat, Mevis, Antego &c... in two books. The
first containing the natural; the second the moral history
of those Islands. Translated by John Davis. J. M. for
Thomas Dring and John Starkey, London.
Rodríguez, Miguel
1997 Religious Beliefs of the Saladoid People. In The
Indigenous People of the Caribbean, edited by Samuel M.
Wilson, pp 80–87. University Press of Florida, Gainesville.
Roe, Peter G.
1989 The Best Enemy is a Defunct, Drilled and Decorative
Enemy: Human Corporeal Art (Frontal Bone Pectorals –
Belt Ornaments, Carved Humeri and Pierced Teeth) in
Precolumbian Puerto Rico. In Proceedings of the Thir-
teenth International Congress for Caribbean Archaeology,
pp. 854–873. Curacao.
Rouse, Irving
1992 The Taínos: Rise and Decline of the People Who
Greeted Columbus. Yale University Press, New Haven.
Royo Guardia, Fernando
1974 El culto de cráneos y los cemíes de algodón entre
antillanos precolombinos. Revista de Arqueología y
Etnología I(II):4–5.
Sauer, Carl Ortwin
1992 The Early Spanish Main. University of California
Press, Berkeley.
Schaffer, William C., Rorbert S. Carr, Jane S. Day, and Michael
P. Pateman
2010 Lucayan-Taíno burials from Preacher ’s Cave,
Eleuthera, Bahamas. International Journal of Osteoar-
chaeology 22:45–69.
Seelye, Elizabeth Eggleston
1892 The Story of Columbus. D. Appleton and Co., New
York.
Stark, Barbara L., Lynette Heller, and Michael A. Ohnersorgen
1998 People with Cloth: Mesoamerican Economic Change
from the Perspective of Cotton in South-central Veracruz.
Latin American Antiquity 9:1–30.
Sued-Badillo, Jalil
1993 La mujer indígena y su sociedad. Editorial Cultural,
San Juan, Puerto Rico.
Symcox, Geoffrey
2002 Repertorium Columbianum: Volume XII: Italian
Reports on America, 1493–1522: Accounts by Contem-
porary Observers. Brepols Publishers, Turnhout, Belgium.
Taylor, Dicey, Marco Biscione, and Peter G. Roe
1997 Epilogue: The Beaded Zemi in the Pigorini Museum.
In Taíno : Pre-Columbian Art and Cult ure from the
Caribbean, edited by Fatima Bercht, Estrellita Brodsky,
John Alan Farmer and Dicey Taylor, pp. 158–168. The
Monacelli Press, New York.
Teerink, B. J.
1991 Atlas and Identification Key: Hair of West-European
Mammals.Cambridge UniversityPress,Cambridge.
Tejera, Emilio
1977 Indigenismos. Editoria de Santo Domingo, Santo
Domingo.
Timyan, Joel
1996 Bwa yo: Important Trees of Haiti. South-Eastern Con-
sortium for International Development, Washington, D.C.
Vallejo, Juan de
1913 Memorial de la vida de Fray Francisco Jiménez de
Cisneros. Imprenta Baily-Bailliere, Madrid.
Vega, Bernardo
1987 [1971] Descubrimiento de la actual localización del
único zemi de algodón antillano aún existente. In Santos,
Shamanes y Zemíes, Fundación Cultural Dominicana,
Santo Domingo.
Walker, Jeffrey B.
1993 , Stone Collars, Elbow Stones and Three-Pointers,
and the Nature of Taíno Ritual and Myth, PhD disserta-
tion, Washington State University, Pullman. University
Microfilms, Ann Arbor.
2005 The Paso del Indio Site, Vega Baja, Puerto Rico: A
Progress Report. In Ancient Borinquen: Archaeology and
Ethnohistory of Native Puerto Rico, edited by Peter E.
Siegel, pp. 55–87. The University of Alabama Press,
Tuscaloosa.
Ostapkowicz and Newsom] “GODS … ADORNED WITH THE EMBROIDERERʼS NEEDLE” 325
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM

326 LATI N AMERICAN ANTIQUITY [Vol. 23, No. 3, 2012]
Wing, Elizabeth S.
2001 Native American Use of Animals in theCaribbean. In
Biogeography of theWest Indies: Patterns and Perspec-
tives, edited by Charles A. Woods and Florence E. Sergile,
pp.481–518. 2nd ed. CRC Press,Boca Raton,Florida.
Yde, Jens
1965 Material Culture of the Waiwái. National Museum of
Copenhagen, Copenhagen.
Notes
1. Kerchache (1994:162) considered the elaborately
beaded cotton cemí in the collections of the Pigorini
Museum, Rome (accession number 4190) a reliquary, but
there is no evidence to suggest that it was used in this way: it
does not contain any human remains, nor is its construction
suggestive of an artifact specifically made to encase bones, as
is the case in the Turin cemí. Instead, the Pigorini cemí con-
sists of two separate cotton elements— a full sized belt at the
base (Biscione 1997:158), and beaded cemí sculpture at the
top, both nailed to a turned wooden base (the latter likely a
historic display support).
2. Animal hair assignments were made using published
data and comparative specimens obtained on loan from the
Florida Museum of Natural History collections, including pri-
mate (Saimuri sciureus, UF#398), felid (Felis wiedii,
UF#6789; Lynx rufus, UF#24401), hutia (Capromys pilorides,
UF#26758; Geocapromys brownii, UF#15249; Plagiodontia
aedium, UF#22399), agouti (Dasyprocta aguti, UF#13304),
guinea pig (Cavia porcellus, UF#9170; Cavia sp., UF#9172),
and bat (Artibeus jamaicensis, UF#2165; Brachyphylla nana,
UF#20704; Eptesicus fuscus, UF#20706; Molossus major,
UF#6549; Monophyllus redmani, UF#31422; Mormoops
blainvillei, UF#13683; Noctilio leporinus, UF#630;
Pteronotus quadridens, UF#13687; Tadarida brasiliensis,
UF#19185). Additional comparative samples utilized in this
analysis and curated in Newsom’s lab include domestic dog
(Canis familiaris), domestic cat (Felis catus), domestic sheep
(Ovis aries), white-tailed deer (Odocoileus virginianus), cot-
tontail rabbit (Sylvilagus sp.).
3. “The women bandage their legs from the calf to the
knee with woven cotton to make them look thicker; they call
this adornment coiro and think it very elegant; they make the
bandage so tight that if it is loosened for any reason the leg
looks very thin. The Jamaicans, both men and women, also
have this custom and even bandage their arms up to the
armpit so that they look as if they wear armlets such as were
once used among us” (Colón 1992:170).
Submitted: April 11, 2011; Revised: January 15, 2012;
Accepted: July 10, 2012.
Delivered by http://saa.metapress.com
Society for American Archaeology - Latin American Antiquity access (804-58-837)
IP Address: 188.220.87.254
Thursday, September 27, 2012 3:29:23 PM