HU Material Culture and Subsistence of Ancient Filipinos in Palawan Essay


Use these 3 sources to write an essay that describes the visual elements and historical elements of the Palawan Burial Caves in the Philippines: Lewis.
“Terminal Pleistocene to Mid-Holocene Occupation and an Early Cremation Burial at Ille Cave, Palawan, Philippines.” Antiquity., vol. 82, 2008.
Szabó, Katherine, and Hazel Ramirez. “Shell-working techniques from the late Neolithic of Palawan Island, the Philippines.” Archaeology in Oceania 53, no. 2 (2018): 90-100.
Adaptation and foraging from the Terminal Pleistocene to the Early Holocene: Excavation at Bubog on Ilin Island, Philippines on JSTOR (
please include information about the artifacts and archaeological findings. I want this essay to include information about the history and culture of the place but at the same time talk about visual elements. For these sources, it includes different caves in Palawan, Philippines. Write a 3 page essay in MLA format. Thanks!
I included files for the sources and here’s the third one’s link:  Adaptation and foraging from the Terminal Pleistocene to the Early Holocene: Excavation at Bubog on Ilin Island, Philippines ( Oceania 44 (2009) 150–159
Worked Shell from Leta Leta Cave, Palawan, Philippines
Keywords: shell artefacts, Palawan, mortuary rituals, Conus, Neolithic, Metal Age
The Leta Leta Cave burial site is a distinctive and enigmatic site of
the Philippine Neolithic, excavated by Robert Fox. Containing a
number of burials, its unusual earthenware pottery – including the
‘yawning-mouth vessel’, small footed goblets and a cut-out
pedestalled bowl – have seen it recognised in the Philippines as an
official site of national significance. In addition to the human
remains and earthenware, Fox recovered a sizeable assemblage of
shell artefacts which, as with other material remains recovered
from Leta Leta, were only cursorily reported in print before his
death. Recent analysis of the Leta Leta worked shell has revealed
the deposition of unfinished as well as finished shell artefacts in
mortuary contexts. As well as giving rare insights into manufacturing protocols in an island where the Neolithic and Metal Age
records are characterised virtually solely by mortuary deposits, the
deposition of unfinished artefacts opens new avenues for the
discussion of Neolithic mortuary practices in Palawan. Results of
a study of the worked shell assemblage, in addition to recent
radiocarbon determinations for the site, are presented here.
The Neolithic and Metal Age archaeological records of
Palawan are amongst the richest in the Philippines, and
being nearly entirely composed of mortuary deposits
provide a valuable window onto ancient Filipino beliefs and
spiritual practices. Much of this material was excavated by
Robert Fox during the 1960s, and while some was reported
upon in print (see especially Fox 1970), Fox’s major
concern was on elucidating the chronological sequence and
broad framework of Palawan prehistory. This necessarily
meant that in-depth considerations of particular sites were
not a feature of his publications. The material excavated by
Fox now provides an outstanding archived resource for
archaeologists investigating the Philippine past, and
specialist studies of various aspects of these assemblages
form important contemporary cornerstones in discussions of
Filipino prehistory (e.g. Dizon et al. 2002; Pawlik and
Ronquillo 2003; Detroit et al. 2004; Hung et al. 2007). The
assemblage from Leta Leta, now curated by the National
Museum of the Philippines in Manila, forms a small fraction
of this archive, and worked shell from this site was studied
in 2002 by the authors. The 2002 study was part of a wider
KS: School of Earth and Environmental Sciences, University
of Wollongong, Wollongong NSW 2522. Email:; HR: Zooarchaeology Section, National
Museum of the Philippines, P. Burgos St., Ermita, Manila
1000, Philippines
comparative project on shell-working technologies across
Island Southeast Asia and the western Pacific (Szabó 2005),
but the results for this important assemblage called for
consideration in their own right.
Background to the site and sample
Leta Leta Cave is a steep fissure in the side of an east-facing
karstic sea-cliff located on Langen Island in the Bacuit
Archipelago off El Nido, northern Palawan, in the
southwestern Philippines (see Figure 1). It is only accessible
from the sea. The floor slopes sharply from the back of the
fissure down to the opening. The treacherous topography of
the site, referred to as ‘difficult and dangerous’ by Fox
(1977: 229), has been aggravated by recent sediment
instability caused by an earthquake in the early 1980s which
effectively destroyed anything that remained of the site.
The site was first recorded as site C-67 by Carl Guthe
during the 1922-25 University of Michigan Expedition to
the Philippines (Solheim 2002: vii, 116), at which time a
Figure 1. Location of Leta Leta and other sites
mentioned in the text.
small surface collection of artefacts was made. These
artefacts, still housed in the University of Michigan
Museum, were incorporated into Wilhelm G. Solheim’s
doctoral work (Solheim 1964, 2002). The site was
completely excavated by Robert Fox and colleagues from
the National Museum of the Philippines in 1965. Due to its
location in northern rather than central Palawan, the site was
not included in Fox’s (1970) monograph on the Tabon
Caves, however numerous references to Leta Leta are made
throughout the volume. A dedicated publication a few years
later provided some further details about the site and
excavations (Fox 1977).
There was no evidence of habitation at Leta Leta and the
site was used primarily, if not exclusively, for burials. Five
burials were excavated including four primary interments
and one secondary ‘bundle’ burial where the bones had been
gathered and placed together in a grave. One of the primary
burials had been interred in a mound of crushed hematite
(Fox 1977: 230) which had coloured both the bones and
surrounding grave goods. There are none of the jar burials
so typical of other later prehistoric burial sites in Palawan
(Fox 1977: 230). As well as highly elaborate, and sometimes
unique, ceramic forms (see Fox 1970: plate 16), a large
number of non-ceramic grave goods were recovered. These
included artefacts in nephrite jade, shale, quartz, jasper,
bone and shell (Fox 1970, 1977). No other classes of
material culture were viewed as a part of this study, but
neither Fox (1970; 1977) nor Hung et al. (2007) make
reference to the presence of unfinished artefacts, and the
photographs of the Leta Leta pottery within Fox (1970:
Plate XVI) show complete and finished vessels. No
information is provided by Fox on the human remains
themselves, and they have not been subjected to further
study. Thus, details of sex, age and treatment of the bodies
are unknown.
Based on the material present in the Guthe collections,
Solheim (1964) originally believed Leta Leta to be a Metal
Age site associated with the Bau ceramic series. Following
excavation by Fox, the lack of any obvious Metal Age
material culture forced a revision of the chronological
estimates, and Leta Leta was recognised as a Neolithic site,
though radiocarbon dates were never obtained. Instead, Fox
offered a relative age based on material affinities with early
Neolithic Tabon Caves assemblages such as Manunggul
Chamber A and Ngipe’t Duldug, thereby dating the bulk of
the Leta Leta remains to 1000–1500 B.C. (Fox 1970: 65, 94,
105-7). He argued the haematite-stained burial (‘Burial 2’)
Lab. Code
Material and
Accession #
Conus litteratus/leopardus
Conus litteratus/leopardus
to be earlier, based on the similarity of both the nature and
placement of the grave goods to the single, primary burial
unearthed at the central Palawan site of Duyong Cave.
However, the date of 4630±250 BP on charcoal argued to be
associated with the Duyong burial (Fox 1970: 17) has been
called into question by Spriggs (1989: 602). Suffice it to say,
there is presently no justification for assigning an earlier
date to Burial 2 at Leta Leta based on associations with
Duyong Cave alone. The lack of an absolute chronology for
Leta Leta has tended to hinder interpretation, and has meant
that the site – despite its clear material importance – has
been omitted from major discussions of Philippine and
Island Southeast Asian Neolithic cultural expressions (e.g.
Spriggs 1989; Bellwood 1997).
The absence of both a site report and Fox’s original
fieldnotes makes the reconstruction of stratigraphy and
associations at Leta Leta challenging. The original artefact
accession records were made available for this study by the
Records Department of the National Museum of the
Philippines, and allow some discussion of artefactual and
stratigraphic associations. While square numbers are
referred to in the accession records, there is no plan that
outlines the grid layout. Depth of deposits is unknown as is
the layout of the burials.
In order to gain some form of chronological control, and
to test Fox’s suppositions regarding a ~1500 BC date for
Leta Leta, two shell samples were selected for conventional
radiocarbon dating. Two Conus litteratus/C. leopardus
blanks with provenance details and clear associations with
significant archaeological features were chosen. Neither
specimen showed signs of post-mortem collection (see
below for a description of taphonomic indicators used to
determine this). The first specimen was recovered from
square 47, 46 cm below the surface inside a red pigmentstained Melo sp. scoop, and returned a date of 3450±80 BP
(ANU-11889, National Museum of the Philippines
accession #65-L-304). The second specimen was recovered
from chamber B1, from ‘subsurface’ deposits associated
with bones and sherds, and returned a date of 3580±70 BP
(ANU-11888, National Museum of the Philippines
accession #65-L-257). Calibrated ages and further information are given in Table 1.
1. Fox’s accession notes make reference to a Chamber A and
Chamber B from the site. Nothing that could clearly be
termed a ‘chamber’ was noted during a visit to the site in
2002, and without the existence of a site plan, Fox’s spatial
reckoning remains unclear.
Square 47, 46cm below
surface inside a Melo sp. scoop
Chamber B, subsurface,
associated with bones and sherds
Age BP
Calibrated Age
BP 1 sigma
3234 – 3425
Calibrated Age
BP 2 sigma
3120 – 3536
3387 – 3554
3321 – 3653
Table 1. Radiocarbon determinations for two shell artefacts from Leta Leta Cave. Calibrated using Calib 5.0.1 with
the marine04 calibration dataset. A delta-R value of 0 has been used, as recommended for the region
(pers. comm. F. Petchy to A. Anderson).
The radiocarbon dates accord very well with Fox’s
original estimates for the site. Without details of
stratigraphy and layout, however, it cannot be determined
whether all the remains at Leta Leta date to this period. The
lack of any metal or glass artefacts makes the presence of
Metal Age deposits unlikely – especially given the
prevalence of glass beads in particular within Metal Age
burial sites in Palawan (e.g. see Francis 2002: Appendix A).
Whether earlier deposits were present is still open to
speculation. While a firm Neolithic date for Leta Leta
certainly sits comfortably with the worked shell types
represented at the site, it further serves to reinforce the
remarkable nature of the earthenware assemblage. As
observed by Solheim (2002: 178), the presence of the
distinctive cut-out pedestalled bowl at Leta Leta demonstrates the antiquity of this form – more commonly
associated with the Metal Age Novaliches pottery complex.
Fragments of such vessels have also been recovered from
9th–10th century AD deposits at Linaminan in central
Palawan (Szabó and Dizon 2007).
Analytical methodology – worked shell
Few standard protocols exist for the analysis of worked shell
and the application of ‘common sense’ mixed with the
transferral of analytical procedures designed to be applied to
other classes of material culture, such as lithic artefacts,
creates much confusion and ambiguity regarding results and
interpretations. The approach taken here was developed as
part of Szabó’s doctoral research (Szabó 2005), and
emphasizes clarity, taphonomy, the recognition and
separation of natural and cultural processes in the
modification of shell, and conservatism in the ascription of
working. This method will not be outlined in detail here, as
only some features are relevant to the Leta Leta worked
shell assemblage.
The analytical procedure for Leta Leta began with the
identification of the raw material of each piece of worked,
or putatively worked, shell. This was done to the lowest
taxonomic level possible based on size, shape, microstructure and surface details such as sculpture and
patterning. Each piece was then checked for the action of
natural processes which would give clues as to whether the
original shell was collected live or post-mortem. Indicators
of post-mortem collection include the muted sculpture and
edge-rounding associated with processes of coastal attrition
or ‘beach-rolling’, the present of adherent organisms such as
barnacles or worm-casts on the inner parts of the shell
(where a mollusc would have resided in life), and the
presence of small holes associated with the action of boring
sponges on the inside of shells or along fracture surfaces.
The latter damage only happens if the shell has been
deposited and spent time in an underwater environment
following death of the mollusc. Evidence for the collection
of empty shells as raw materials indicates that particular
shells were sought or deliberately selected independent of
any selection procedures for shells collected as part of a
subsistence strategy. Post-mortem changes in shell structure
and composition further mean that the structural properties
of the raw material itself will be slightly different – most
significantly in the loss or reduction of the organic
components of the matrix.
Clear signs of human modification of shells with the
intent of producing an artefact include grinding,
cutting/sawing, freehand abrasion, etching/scoring with a
point and drilling. Straight edges and flat surfaces are
typically not enough to directly infer human modification,
as, say, splitting along natural growth lines can produce a
very straight break and agents such as resident hermit crabs
can produce isolated patches of wear on shells. Carnivorous
gastropods within the Muricidae (rock shells) and Naticidae
(moon snails) access their prey by drilling through the shell
of the victim with a boring accessory combined with acid,
thus leaving a perfectly round hole often confused with
human modification by drilling.
Given these issues, the identification of striations
associated with grinding and cutting, use-wear on artefact
surfaces and perforations, and bevelling of cut edges can
greatly assist in separating natural from cultural
modification. Direct and indirect percussion are more
difficult to pinpoint as the non-homogenous structure of
shell tends not to accommodate distinctive traces of impact
such as a bulb of percussion or associated ripples. It should
be noted that the identification of natural processes of
modification does not rule out either the human collection of
the shell or its further modification or use. For example,
naturally-perforated shells may be used as beads or
pendants. In such situations, the identification of use-wear
or the combination of natural and cultural processes of
modification is important.
Leta Leta was one of the few Philippine worked shell
assemblages studied which contained unfinished artefacts,
and thus had the potential to inform upon technological
choices and working protocols. Being a burial site rather
than a workshop it was unlikely that all elements of any
given reduction sequence were present, and thus the
linkages between pieces of worked shell at different stages
within the sequence had to be inferred. This was done
through grouping together those artefacts, blanks and
preforms produced in particular species/taxa, and assessing
levels of reduction. As will be seen in more detail
below, it was apparent with Conus spp. shells that large
whole shells were initially reduced by direct percussion,
followed by grinding of the opposing spire and fractured
body surfaces. Detached but unground spires are referred
to here as ‘blanks’, while partially ground spires are referred
to as ‘preforms’. One or more perforations were then
generated to create a shell disc. In other less wellrepresented taxa, comments could be made regarding
working techniques, but a full reduction sequence could not
be reconstructed.
What follows below is, initially, an assessment of raw
material selection and shell working evidenced at Leta Leta.
Following this is a discussion of patterns of deposition as it
relates to a burial site context.
The Leta Leta worked shell assemblage
All shell recovered from Leta Leta by Fox was recognised
as being worked and shell representing subsistence refuse
(midden) would seem to be completely absent. As the site
literally drops straight down into the South China Sea and
live mollusc resources are clearly available, the lack of
midden reinforces the site’s strictly burial function. The Leta
Leta deposits are rich in worked shell with several hundred
artefacts dominated by beads and modified Conus spp.
spires. Although a restricted number of artefact types and
raw materials dominate, the site contains greater diversity,
as well as a greater sample of worked shell overall, than any
other site in Palawan with the exception of Ille Cave and
Shelter – also in northern Palawan (see Szabó 2005; Szabó,
Swete Kelly and Peñalosa 2004). In his writings on the site,
Fox (Fox 1970, 1977) mentions Conus spp. rings and discs,
shell beads including ‘sequin-shaped shell discs’ and
modified Cypraea and Nassarius, ‘scoops’ of Melo
amphora, Turbo marmoratus and Nautilus, rings in Trochus
and Patella (limpet), and pendants of Haliotis (abalone),
Cypraea and Strombus. In her survey of modified shell and
bone artefacts from Philippine sites, Alba (1998: 52) also
mentions the presence of a number of pierced opercula of
the terrestrial snail Opisthoporus quadrasi (Cyclophoridae).
558 Beads made from the spire of a Strombus shell.
Many covered in a thin dusting of red pigment.
208 The dorsa have been knocked away from the
aperture/parietal callous. Most are worn and chalky.
Chamber B
Subsurface; recovered
during screening.
Shell Rings
Conus litteratus/
On fore-arm of skeleton in this square.
Conus sp.
On rock shelf.
Conus sp.
0–20cm below surface
Conus spp. discs
Conus sp.
Ground spire with two perforations. Diameter 29.07 mm.
Conus sp.
Conus sp.
Conus sp.
Grave 5 in association with skull
Conus sp.
Conus sp.
Accession record not in accordance
with artefact
No accession numbers
Conus sp.
From an extremely large shell. Worn and eroded with a
single perforation by the aperture. Diameter 90.08 mm.
Highly eroded and chalky with a central perforation.
Diameter 40.83 mm.
Worn and ground spire with a ‘scratched’ perforation
near the aperture. Diameter 37.66 mm.
A water-rolled spire, modified after collection. A hole
has been drilled near the margin. Diameter 48.52 mm.
Ground and perforated spires. Four are perforated at the
centre and four near the aperture. Diameters range from
27.23 mm–64.52 mm.
Ground spire – non-perforated. Diameter 39.24 mm.
Accession record not in accordance
with artefact
Accession record not in accordance
with artefact
35 cm below surface
Shell Beads
Strombus canarium
Nassarius arcularis
and/or N. pullus
It is possible that these equate to Fox’s ‘sequin-shaped
shell discs’.
The National Museum of the Philippines made a sample
of worked shell from Leta Leta Cave (n = 815) available for
study by the authors. The analysis was undertaken at the
National Museum in Manila. Worked shell taxa in the
analysed sample included Conus litteratus and/or Conus
leopardus, Strombus spp. and Nassarius spp. (see Table 2).
Forty-nine Conus spp. artefacts were analysed, and where
condition and degree of working allowed species
identification, it was apparent that all were manufactured
from C. litteratus and/or C. leopardus. These two species
are among the largest species of Conus extant in the tropical
Indo-Pacific. They are both patterned with dark rows of
spots against a white background and are found in similar
weedy sand and rubble environments. The only way of
visually separating the shells of the two species is the
presence of a patch of dark coloration near the siphonal
canal of C. litteratus which is absent in C. leopardus. If this
part of the shell is not present – as is common in worked
shell assemblages where the body is the first part to be
removed and discarded – then the two species cannot be
separated. Within the Leta Leta assemblage, specimens at
different stages of working indicate that shells used as raw
materials were sometimes collected post-mortem. The
Complete broad rings, each with a single perforation by
the aperture. Internal diameters range from 50–60 mm.
Widths range from 23.5–36 mm.
Complete narrow ring. Highly polished through use.
Internal diameter 40 mm. Max. width 8.12 mm.
Complete narrow ring. Highly polished through use.
Domed plano-convex cross-section. Internal diameter
60 mm. Max. width 8.13 mm.
No accession number
continued overleaf
N Description
Conus spp. spire ornament preforms
Conus litteratus/
Detached spire, partially ground. The outer spire surface
is covered in marine growths; probably collected
post-mortem. Diameter 40.89 mm.
Conus litteratus/
Detached spire, partially ground. Diameter 43.91 mm.
Conus litteratus/
Detached spire, partially ground. Diameter 33.81 mm.
Conus sp.
Heavily eroded detached and partially ground spire.
Diameter 43.91 mm.
Conus litteratus/
Detached spires, partially ground. Diameters: 28.38 mm
and 34.19 mm.
Conus litteratus/
Detached and partially ground spires. Diameters range
from 38.98 mm to 45.84 mm.
Conus spp. spire ornament blanks
Conus litteratus/
Detached spire and posterior part of body whorl.
Diameter 37 mm.
Conus litteratus/
Detached spire and posterior part of body whorl.
Traces of powdered red pigment. Diameter 59.50mm.
Conus litteratus/
Spire and posterior part of body whorl detached through
successive percussive blows to the body whorl.
Diameters: 36.92 mm and 59.32 mm.
Conus litteratus/
Spire detached by successive blows to the body whorl.
Diameter 46.63 mm.
Conus litteratus/
Spire detached by successive blows to the body whorl.
Diameter 28.43 mm.
Conus litteratus/
Whole spire smooth and well rounded; water-rolled.
Diameter 52.76 mm.
Conus litteratus/
Spire detached by successive blows to the body whorl.
Diameter 39.92 mm.
Conus litteratus/
Spire detached by successive blows to the body whorl.
Traces of powdered red pigment coating. From a large
deformed senile specimen. Diameter 58.32 mm.
Conus litteratus/
Spire detached by successive blows to the body whorl.
Traces of powdered red pigment coating.
Diameter 60.94 mm.
Spire detached by successive blows to the body whorl.
Conus sp.
Diameter 26.77 mm.
Conus sp.
Bleached and sponge-eroded; collected post-mortem.
Detached spire. Diameter 28.89 mm.
Spires detached by successive blows to the body whorl.
Conus litteratus/
Diameters range from 29.54 mm–48.22 mm.
Chamber B
37cm below surface
Chamber B
Squares 47 & 53 in grave
Chamber B
46cm below surface
No accession numbers
Chamber B
46cm inside a Melo sp. scoop
Chamber B
Shelf B1
47 cm below surface
35 cm below surface
35 cm below surface
36 cm below surface inside a
Melo sp. scoop
46 cm below surface
Chamber B
Accession record not in accordance
with artefact
No accession numbers
Table 2. Worked shell from Leta Leta included within this study.
Samples indicated by ** are the specimens selected for radiocarbon dating (Table 1).
specimen shown in Figure 2a has been water-rolled,
producing a smooth surface that was then minimally ground
on the outer spire face. The specimen shown in Figure 2b
shows extensive damage by boring sponges and it is likely
that this too indicates post-mortem collection.
Two formal artefact types produced in C. litteratus/
C. leopardus were present in the analysed assemblage:
perforated discs (n = 13) and both narrow (n = 2) and broad
(n = 3) rings/’armbands’. In addition to finished artefacts, a
number of blanks (n = 20) and preforms (n = 11) were also
present. Specimens have been defined as ‘preforms’ where
grinding has begun and ‘blanks’ where there is evidence of
percussion but no grinding. The Conus spp. disc blanks –
which are generally smaller than those destined to become
rings – indicate that the body and spire of the shell were
separated using direct percussion (Figure 2c and d). After
this, the body whorl edge was ground flat (Figure 3a), and
then the spire was ground flat (Figure 3b). A perforation was
then generated near the aperture or through the centre. There
seem to have been a variety of ways of effecting the
perforation including drilling (Figure 3c and 2a
aperture/edge perforations), grinding (Figure 3d central
perforation) or ‘scratching’ with a sharp point (Figure 3d).
Five finished and complete Conus spp. rings from Leta
Leta were analysed. Of the two narrow rings, one was
apparently finished and worn despite an irregularly-shaped
internal surface (Figure 4a). The second was well abraded
on all surfaces, finished and worn (Figure 4b). The three
broad rings were recovered from a fore-arm of burial #1
(Figures 4c–e). All three specimens have a drilled
perforation near the aperture. While such perforations could
be interpreted as being for suspension of the artefact as a
pendant, this seems not to be the case here given the
positioning of the rings on the arm of the body. Rather, it
seems that other items/artefacts were suspended from this
hole as indicated by differential wear around the perimeter.
No debitage or preforms obviously relating to ring
manufacture were contained within the Leta Leta sample.
Figure 2. Conus spp. working at Leta Leta; (a) water-rolled
and then minimally ground and perforated Conus sp. spire,
no provenance details; (b) detached spire of a Conus sp.
shell showing extensive damage from boring sponges, no
provenance details; (c) spire of a Conus litteratus/
C. leopardus detached by direct percussion, square 53,
35cm below surface (accession #65-L-311); (d) spire of
a Conus litteratus/C. leopardus detached by direct
percussion, no provenance details.
Scale bars in centimetres.
Figure 3. Conus spp. working at Leta Leta – the
production of perforated Conus spire discs; (a) the body
whorl edge of the blank is ground down, Chamber B,
subsurface (accession #65-L-257); (b) Conus litteratus/
C. leopardus spire partially ground on both faces,
subsurface (accession #65-L-222); (c) Ground and
perforated Conus sp. spire with a drilled perforation at the
perimeter. The central perforation is deteriorated and
method of production cannot be deduced, no provenance;
(d) ground and perforated Conus sp. spire with a scratched
perforation at the perimeter, grave 5 (square 52) near skull
(accession #65-L-340). Scale bar in centimetres.
Figure 4. Conus spp. rings at Leta Leta; (a) narrow Conus
sp. ring, on top of rock shelf in square 18a (accession
65-L-121); (b) narrow Conus sp. ring, square 24 0-20cm
(accession 65-L-16); (c) broad Conus sp. ring with
perforation at aperture, square 18 on arm of burial #1;
(d) broad Conus sp. ring with perforation at aperture,
square 18 on arm of burial #1; (e) broad Conus sp. ring
with perforation at aperture, square 18 on arm of burial #1.
Scale bars in centimetres.
Many hundreds of shell beads were recovered during the
excavation of Leta Leta Cave. Fox (1977: 229-30) outlines
how the many tiny beads that fell through the cracks of the
platforms constructed for the excavation were later screened
from sediments below. The beads made of pierced
Opisthoporus quadrasi opercula mentioned by Alba (1998:
52), as well as the small Cypraea spp. beads mentioned by
Fox (1977: 233), were not part of the analysed sample. Of
the remaining bead types cited for Leta Leta, 558 of the
‘round, thin, and flat shell discs’ mentioned by Fox (1977:
233) were individually analysed, and a reconstructed strand
of Nassarius spp. beads (n = 208) from the public display
gallery of the National Museum was also studied, though
not unstrung.
Although disc beads are generally regarded to be manufactured from small Conus spp. spires (Alba 1998: 52-3),
the raw material for such artefacts at Leta Leta Cave has
been identified here as Strombus spp. The evidence for the
manufacture of these beads, present at nearby Ille Cave,
indicates that spires of Strombus canarium and probably
Strombus luhuanus were utilised in manufacture. Strombus
spp. beads, as compared to those manufactured in Conus
spp., tend to have different whorl spacing and the perimeter
of the bead can be observed to be sloping in profile rather
than perpendicular to the face of the bead. Based on
evidence from Ille Cave (Szabó 2005), the perimeters of
Strombus spp. beads are left unground. All the Leta Leta
examples have been ground on both faces. The perforation
was formed as a consequence of grinding the conical spire
rather than being drilled. Five examples are illustrated in
Figure 5a-e. Figure 5c shows clearly the ground perforation,
where, due to uneven grinding of the preform, the hole is at
the apex of the spire rather than the centre of the bead.
Species of the Nassariidae chosen for bead manufacture
include Nassarius arcularius and Nassarius pullus. The
dorsum of each shell has been removed by percussion to
enable threading or attachment. Specimens are unground
Figure 5. Ground beads manufactured from Strombus spp.
from Leta Leta cave, screened from sediments below
excavation platforms. Scale bar in centimetres.
but use-wear traces on identical beads from Ille Cave (Szabó
2005) suggest that grinding was not a usual part of the
production process.
Leta Leta worked shell in context
While the evidence for Conus spp. disc production at Leta
Leta is regionally-rare and consequently illuminating, it is
intriguing that this evidence occurs at Leta Leta at all. As
stated above, the site is unstable, accessible only by sea, and
contains no evidence for habitation. Given this, the presence
of unfinished artefacts was unexpected. The lack of Conus
spp. body fragments suggests that shell-working did not
take place at the site. Rather, blanks and preforms appear to
have been deposited as grave goods along with finished
While numerically Leta Leta would appear to have the
largest number of unfinished artefacts deposited within
burial contexts in Palawan, the practice is not without
precedent. During the same season in which Leta Leta was
excavated, Fox excavated another cave site on Langen
Island called Paredes Shelter (see Fox 1966, 1970). The
Paredes site consists of a small, stratified shelter and three,
unstratified, historic period ‘grottos’ (Fox 1970: 176). The
stratified shelter was excavated by Fox, and was found to
contain evidence of three separate periods of use. The
uppermost/surface layer of the shelter included three
shroud-wrapped burials with associated material dated
tentatively by Fox to the 18th century (Fox 1970: 176). A jar
burial assemblage was found below this, and the presence of
a bronze axe and jade indicated Metal Age deposition (Fox
1970: 177). A third layer below this revealed three, primary
Neolithic burials with associated artefacts of Conus spp. and
a Tridacna sp. adze (Fox 1970: 177). Fox (1970: 177)
considered these shell artefact types to indicate a cultural
association with Neolithic levels at Duyong Cave and thus
dated the lower layer to the ‘early Neolithic’. No
radiocarbon determinations were ever obtained for the site.
Two modified Melo spp. shells were noted by Fox in the
accession records, both being found together and bearing
the same accession number. They were both recovered at a
level above 50 cm in depth, and therefore have a probable
Metal Age association. One of these artefacts was analysed
by the authors in 2002 (see Figure 6), and while Fox calls
this artefact a ‘scoop’ in the accession records, it is perhaps
more correct to term it an unfinished artefact or ‘reduced
shell’. Unlike finished examples of Melo artefacts
elsewhere, none of the edges have been trimmed or abraded.
The inner whorls have been removed by a process of
chipping and snapping leaving only the body whorl and the
spire. Such techniques employed in initial stages of working
have been noted for Ille Cave and Shelter (see Szabó 2005:
260). If this indeed represents an unfinished artefact utilised
as a grave good, it groups well with Conus spp. blanks and
preforms recovered from nearby Leta Leta.
A further site excavated by Fox and colleagues as a part
of the Tabon Caves project in central Palawan was Batu Puti
Figure 6. Reduced Melo broderipii shell from the Paredes
Shelter burial deposits. Recovered from square J,
subsurface (accession #65-W-I-18). Scale bar in
Figure 7. Trochus niloticus ring preform of probable Metal
Age association from Batu Puti Cave, square 4, 60cm
below surface. Scale bar in centimetres.
Cave. In his 1970 monograph on the excavations Fox barely
mentions the site, and we have found no references to the
site at all in unpublished material held by the National
Museum of the Philippines. From scattered comments in
The Tabon Caves, it is apparent that the site contained both
primary and jar burials (Fox 1970: 9). Fox (1970: 118) links
the jar burials with the ‘developed’ Metal Age, based on the
noted similarity in material culture to Manunggul Chamber
B, which Fox reports to have a radiocarbon determination
of 190 BC (Fox 1970: 118-9). The primary burial(s) are
apparently Neolithic, with ground stone and Tridacna spp.
adzes being recovered (Fox 1970: 9, 62). Fox notes that the
assemblage was ‘badly disturbed’ (Fox 1970: 62).
The accession records for Batu Puti list several hundred
shell artefacts; mainly shell ‘ornaments’, beads, spoons, and
rings. Fox (1970: 62, 118, 146) mentions a few of these
including Tridacna sp. adzes, Anadara sp. ‘lime containers’,
Nautilus sp. spoons, generic ‘shell beads’ and a scoop
manufactured from Cassis cornuta. A small sample of
artefacts from Batu Puti studied by the authors included four
fragments of a ring preform manufactured from Trochus
niloticus which were able to be reconstructed to form a nearcomplete example (Figure 7). The preform was made from
the widest part of the shell where the ventral surface and
body whorl meet. The inner architecture of the shell has
been chipped away with a sharp point. The body whorl edge
has been ground, and it is uncertain whether this edge was
originally cut or chipped. The ventral surface has been
ground flat. The aperture area of the body/ring is not
present, so it is unclear if or how the ring connected to form
a complete circle. The preform is not directly associated
with iron or glass, however a black and white glass ‘eye
bead’ was recovered from 5 cm above in the same square.
From these few known occurrences of the deposition of
unfinished shell artefacts with burials, some noteworthy
conclusions can be drawn. Firstly, this practice clearly spans
from the Neolithic (Leta Leta) into the Metal Age (Paredes
Shelter), with the chronological association at Batu Puti
being somewhat ambiguous. Secondly, there is no clear
association between the practice of depositing unfinished
shell artefacts and any particular mode of burial. At Leta
Leta, blanks and preforms are in association with primary
burials. An association between unfinished artefacts and the
Leta Leta secondary bundle burial cannot be established
based on the accession records or other available
information, but red pigment on a number of shell artefacts
would suggest an association between various artefacts and
the pigmented primary burial. At Paredes Shelter it would
appear that the worked Melo sp. shells are in association
with jar burials, and again, the case at Batu Puti is
ambiguous, but is either an association with primary
interments or jar burials. Thirdly, it is apparent that the
deposition of unfinished shell artefacts with burials is not a
practice restricted to northern Palawan, with the Batu Puti
example testifying to its occurrence in the Tabon area.
Linkages between northern and central Palawan are also
evident in the distribution of shell bead types. Strombus spp.
beads are ubiquitous at both Leta Leta and Ille Cave, with
several hundred having been recovered from the latter
(Szabó 2005: 256-258). In central Palawan, Chazine (n.d.)
includes images of a number of Strombus spp. disc beads –
although misidentified as Conus sp. – from the ‘Upper
Duyong’ site, near Fox’s original Duyong Cave (see Fox
1970). Nassarius spp. beads have likewise been recovered
in some numbers from Ille Cave (Szabó 2005: 258-259).
In central Palawan, Fox reported Nassarius spp. beads from
the Neolithic jar burial site Ngipe’t Duldug (Fox 1970: 107)
and Metal Age jar burial site Pagayona Cave (Fox
1970: 147).
One element of the Leta Leta worked shell assemblage
which is, so far, unique in Palawan, is the presence of broad
C. litteratus/C. leopardus rings, with a distinctive perforation by the aperture. Reconstruction of the production
protocols for broad and narrow Conus spp. rings recovered
from Lapita contexts in the western Pacific has
demonstrated that there is greater risk and labour involved
in the production of broad rings (Szabó 2005: 169-170). As
a consequence, broad Conus spp. rings tend to be both less
common, and often intensively curated (Szabó 2006; Szabó
and Summerhayes 2002: 96). While there is no evidence of
the curation of broken Conus spp. broad rings in the
Philippines, the mere fact that a substantial portion of the
body of the shell must be preserved intact during initial
reduction necessitates greater care than is required for
producing narrow rings. Conus spp. narrow rings, on the
other hand, are not uncommon and seemingly widespread,
with small numbers having been recovered from range of
sites including Ille Cave (Szabó 2005: 262), Sa’gung Shelter
(Szabó 2005: 272-273) and Batu Puti (Szabó 2005: 276277). These sites have mixed Neolithic and Metal Age
deposits excepting Sa’gung, where published and archived
information on stratigraphy and artefact provenances is
insufficient to ascribe artefacts to particular layers. Thus,
ascription of artefacts from these three sites to either the
Neolithic or the Metal Age is presently not possible.
Leta Leta Cave is unique and still rather enigmatic within
the cache of Philippine archaeological sites. Both the lack of
a firm chronology and details about excavation have, and
continue to, hinder interpretation. Despite this, valuable
information can still be extracted and a number of important
conclusions drawn. One of the most interesting of these is
the clear strands of continuity from the Palawan Neolithic
into the Metal Age. This boundary is often seen as being
rather sharp, with attendant concepts such as ‘Indianisation’
and the rising tide of new trade goods being seen to displace
locally-produced material culture (e.g. Francis 2002:
Appendix A). In fact there seems to be rather little evidence
of such a disjuncture, with all of the shell artefact types seen
at Leta Leta having both Neolithic and pre-AD 1000 Metal
Age associations throughout Palawan. This observation is
not restricted to shell, with the continuing occurrence and
valuing of polished stone adzes witnessed at the Linaminan
open site in central Palawan well into the Metal Age (Szabó
and Dizon 2007). Earthenware vessels such as the Leta Leta
cut-out pedestalled bowl provide another thread of
continuity over the Neolithic/Metal Age threshold. It would
appear that, while foreign trade goods quickly came to
occupy an important place in local Palawan economy and
consciousness, this did not result in a simple displacement
of extant items of use and value.
While no clear lines between Neolithic and early Metal
Age shell-working can be drawn in Palawan, so it is that
there are no apparent distinct associations between
particular artefacts and differing modes of burial. Whether
secondary jar burial or extended primary interment, the
same roster of shell artefacts tend to be found in association.
This begs the question of the significance of differing modes
of burial, and certainly casts doubt upon any interpretation
that would seek to ascribe different burial modes to separate
socio-cultural groups (see discussion in Fox 1970: chapter
VII). Arguments for a simple chronological partitioning of
burial traditions also falter when sites such as Leta Leta,
Ngipe’t Duldug and the middle (Neolithic) and upper (Metal
Age) layers of Duyong Cave reveal a mixture of jar burials,
primary extended inhumations and secondary interments
with overlapping artefact types and, in all likelihood,
There has been a tendency within literature focused upon
shell ornaments to see such artefacts as ‘prestige goods’
where value is often attached to not only the finished
artefact, but the hands through which it has passed in
hypothesized exchange networks (e.g. see Kirch 1991;
Trubitt 2003). The production process itself is rarely the
focus of interpretational significance or factored into
assessments of ‘value’. The deposition of unfinished Conus
shell artefacts within mortuary contexts at Leta Leta, as well
as the deposition of other types of unfinished shell artefacts
within other mortuary sites in Palawan, clearly suggests that
meaning did not wholly reside in finished artefacts or their
associated genealogies. The deposition of pigmented Conus
sp. debitage within mortuary contexts at the Niah Caves, on
the neighbouring island of Borneo (Szabó, unpublished
data), further implies that such patterning is not restricted
to Palawan. Certainly, any attempt to comprehend the
deposition of unfinished shell artefacts and debitage within
burial sites requires a mental shift away from analogies with
ethnographic exchange networks (e.g. Malinowski 1922)
which have so influenced interpretative thinking.
There is currently little indication of exchange of marine
shell objects in the Neolithic (or Metal Age) Philippines,
although the paucity of habitation and/or manufacturing
sites precludes any balanced or in-depth consideration. The
overlap in shell artefact types between the majority of
Palawan Neolithic and Metal Age sites may just as likely
indicate shared local traditions of production as a trade in
finished artefacts. However, it can be hypothesized that the
deposition of unfinished artefacts within mortuary contexts
implies local production at or near the sites concerned.
Exactly what the deposition of unfinished shell artefacts
in burials ‘means’ is difficult to say, and a variety of
interpretations is possible. Such potential interpretations
include unfinished artefacts ‘standing in for’ finished ones,
blanks and preforms indicating the significance of the
production process, or perhaps the significance lies in the
worked raw material itself. While it is tempting to speculate
upon the various possibilities, and draw upon the rich stock
of regional ethnographic literature, we hesitate to
overextend interpretation when too little is known of
Neolithic Filipino social structure and differentiation and
the spatial and social dynamics of craft production. In order
to investigate the various possibilities in greater depth, more
detailed analytical and contextual data would be required,
such as age and sex data of the associated skeletons, and
associations with other forms of material culture. What can
be currently stated with certainty is that shell ornaments
generally, and Conus spp. artefacts in particular, are clearly
of value. Not only do they consistently occur within
Palawan mortuary contexts (see Fox 1970; Szabó 2005),
their significance clearly extends beyond the isolated
artefact itself. The location and excavation of Neolithic nonmortuary sites within Palawan would be of great value in
efforts to clarify the structure, process and meaning of not
only shell artefacts, but local Neolithic lifeways at large.
Thank you to the National Museum of the Philippines,
particularly Ms Corazon Alvina (Director) and Mr Wilfredo
Ronquillo (Chief, Archaeology Division) for allowing and
supporting the research presented here. Mr Ronquillo and
Dr Eusebio Dizon gave permission for samples to be
removed from the collections for radiocarbon dating. Ms
Amalia de la Torre went out of her way to provide accession
records relating to Leta Leta Cave, Paredes Shelter and Batu
Puti Cave, and we are grateful for these additional original
data. Thanks also go to Mary Beth Trubitt, Peter White and
an anonymous reviewer for comments on a previous draft.
Radiocarbon dates were funded by the Centre for
Archaeological Research, Australian National University.
Alba, E.D.V. 1998. Modified Bones and Shells from Philippine
Archaeological Sites. National Museum Papers (Manila,
National Museum of the Philippines) 8(1): 45-80.
Bellwood, P. 1997. Prehistory of the Indo-Malaysian Archipelago.
Revised edition. Honolulu: University of Hawaii Press.
Chazine, J-M. n.d. Archaeological Investigations Program,
Palawan – Philippines. Unpublished manuscript. Quezon City:
Archaeological Studies Program, University of the Philippines.
Détroit, F., E. Dizon, C. Falguères, S. Hameau, W. Ronquillo and
F. Sémah 2004. Upper Pleistocene Homo sapiens from the
Tabon Cave (Palawan, the Philippines): description and dating
on new discoveries. Comptes Rendus Palevol 3: 705-712.
Dizon, E., F. Détroit, F. Sémah, C. Falguères, S. Hameau, W.
Ronquillo and E. Cabanis 2002. Notes on the morphology and
age of the Tabon Cave fossil Homo sapiens. Current
Anthropology 43(4): 660-666.
Fox, R.B. 1966. First Progress Report to the National Geographic
Society of the Explorations and Excavations of Cave Sites of
Ancient Man on Palawan Island, Philippines. Unpublished
manuscript. Manila: National Museum of the Philippines.
–––– 1970. The Tabon Caves. Manila: National Museum
–––– 1977. “Leta-Leta Cave – The science and art of cave
archaeology,” in Filipino Heritage: The Making of a Nation.
Vol. 1. Edited by A.R. Roces, pp. 228-34. Manila: Lahing
Pilipino Publishing Inc.
Francis, P. Jr. 2002. Asia’s Maritime Bead Trade: 300 BC to the
Present. Honolulu: University of Hawaii Press.
Hung, Hsiao-Chun, Yoshiyuki Iizuka, Peter Bellwood, Kim Dung
Nguyen, Bérénice Bellina, Praon Silapanth, Eusebio Dizon,
Rey Santiago, Ipoi Datan, and Jonathan H. Manton 2007.
Ancient jades map 3,000 years of prehistoric exchange in
Southeast Asia. Proceedings of the National Academy of
Sciences 104(50): 19745-19750.
Kirch, Patrick V. 1991. Prehistoric Exchange in Western
Melanesia. Annual Review of Anthropology 20: 141-165.
Malinowski, B. 1922. Argonauts of the Western Pacific. London:
Routledge and Kegan Paul.
Pawlik, A.F. and W.P. Ronquillo 2003. The Palaeolithic in the
Philippines. Lithic Technology 28(2): 79-93.
Solheim, W.G. II 1964. The Archaeology of the Central
Philippines: A Study Chiefly of the Iron Age and its
Relationships. Manila: Monographs of the National Institute of
Science and Technology, no. 10.
–––– 2002. The Archaeology of the Central Philippines: A study
chiefly of the iron age and its relationships. Revised edition.
Manila: Archaeological Studies Program, University of the
Spriggs, M. 1989. The dating of the Island Southeast Asian
Neolithic: an attempt at chronometric hygiene and linguistic
correlation. Antiquity 63: 587-613.
Szabó, K. 2005. Technique and practice: Shell-working in the
western Pacific and Island Southeast Asia. Doctoral thesis.
Australian National University, Canberra.
–––– 2006. Shell artefacts from the Lapita site of Bourewa, Viti
Levu, Fiji. Unpublished manuscript. Suva: University of the
South Pacific.
Szabó, K. and G. Summerhayes 2002. “Lapita worked shell
artefacts: new data from early Lapita,” in Fifty Years in the
Field: Essays in Honour and Celebration of Richard Shutler
Jr’s Archaeological Career. Edited by C. Sand, S. Bedford and
D. Burley, pp. 91-100. Auckland: New Zealand Archaeological
Association Monograph 25.
Szabó, K., M.C. Swete Kelly and A. Peñalosa 2004. “Preliminary
results from excavations in the eastern mouth of Ille Cave,
northern Palawan,” in Southeast Asian Archaeology: A
festschrift for W.G. Solheim II, Edited by V. Paz, pp. 209-224.
Manila: University of the Philippines Press.
Szabó, K. and E. Dizon 2007. The archaeology of Linaminan,
central Palawan: A preliminary report on excavations. Hukay
11: 1-84.
Trubitt, Mary Beth D. 2003. The Production and Exchange of
Marine Shell Prestige Goods. Journal of Archaeological
Research 11(3): 243-277.

Purchase answer to see full

We offer the bestcustom writing paper services. We have done this question before, we can also do it for you.

Why Choose Us

  • 100% non-plagiarized Papers
  • 24/7 /365 Service Available
  • Affordable Prices
  • Any Paper, Urgency, and Subject
  • Will complete your papers in 6 hours
  • On-time Delivery
  • Money-back and Privacy guarantees
  • Unlimited Amendments upon request
  • Satisfaction guarantee

How it Works

  • Click on the “Place Order” tab at the top menu or “Order Now” icon at the bottom and a new page will appear with an order form to be filled.
  • Fill in your paper’s requirements in the "PAPER DETAILS" section.
  • Fill in your paper’s academic level, deadline, and the required number of pages from the drop-down menus.
  • Click “CREATE ACCOUNT & SIGN IN” to enter your registration details and get an account with us for record-keeping and then, click on “PROCEED TO CHECKOUT” at the bottom of the page.
  • From there, the payment sections will show, follow the guided payment process and your order will be available for our writing team to work on it.