Terminal Pleistocene to mid-Holocene occupation and an early cremation burial at Ille Cave, Palawan, Philippines

Terminal Pleistocene to mid-Holoceneoccupation and an early cremationburial at Ille Cave, Palawan, PhilippinesHelen Lewis1, Victor Paz2, Myra Lara2, Huw Barton3, Philip Piper2,Janine Ochoa2, Timothy Vitales2, A. Jane Carlos2, Tom Higham4,Leee Neri2, Vito Hernandez2, Janelle Stevenson5, Emil CharlesRobles2, Andrea Ragragio2, Rojo Padilla2, Wilhelm Solheim II2 &Wilfredo Ronquillo6

Excavations at a cave site on the island of Palawan in the Philippines show occupation from c.11000 BP. A fine assemblage of tools and faunal remains shows the reliance of hunter-foragersswitching from deer to pig. In 9500-9000 BP, a human cremation burial in a container wasemplaced, the earliest yet known in the region.

Keywords: Southeast Asia, Philippines, Palawan, Ille Cave, Late Palaeolithic, terminalPleistocene, early to mid-Holocene, cremation, occupation, radiocarbon dates

IntroductionNew excavations at Ille Cave, Palawan, Philippines have produced the first well-stratified anddated terminal Pleistocene to mid-Holocene archaeological sequence from the Philippinearchipelago. This sequence includes evidence of changing subsistence and settlementpatterns, as well as an early cremation burial. The island of Palawan, situated betweenBorneo to the south and the Philippine archipelago to the north and east (Figure 1), isimportant for its biogeographic links with the Sundaic region of Southeast Asia (e.g. Heaney1985; Heaney et al. 2005; Merill 1923; Reis & Garong 2001), and for interpreted patternsof ancient human migration between, and cultural connections with, mainland SoutheastAsia, the rest of the Philippines, Taiwan, Borneo and Indonesia (Bellwood 1997; 2005;O’Connor 2007; Solheim 2006). Many cave sites are known from the island, includingTabon Cave, which has the earliest recovered human remains in the Philippines (c . 45kya)(Detroit et al. 2002; Dizon 2003; Fox 1970), and some of the earliest in Island Southeast

1 School of Archaeology, University College Dublin, Belfield, Dublin 4, Ireland2 Archaeological Studies Program, University of the Philippines Diliman, Quezon City, Manila, Philippines3 School of Archaeology and Ancient History, University of Leicester, Leicester, UK4 Oxford Radiocarbon Accelerator Unit, RLAHA, University of Oxford, Oxford, UK5 Research School of Pacific and Asian Studies, Australian National University, Canberra, Australia6 Archaeology Division, National Museum of the Philippines, Manila, Philippines

Received: 10 May 2007; Accepted: 13 September 2007; Revised: 30 October 2007

antiquity 82 (2008): 318–335

318

Res

earc

h

Helen Lewis et al.

Figure 1. Location of Ille Cave and other sites mentioned in the text (H. Lewis).

Asia (e.g. O’Connor 2007). There are also several important Neolithic and Metal Age cavecemeteries in Palawan (Fox 1970).

Ille is a solution cave and rockshelter at the base of a c . 75m-high karst tower nearthe village of New Ibajay, El Nido, north Palawan (119◦30′19′′E, 11◦11′46′′N). The sitecomprises east and west rockshelter mouths, opening onto a relatively flat platform of siltloam (Figures 2 and 3), in a setting of light woodland. From the Ille summit other localtowers are visible, and 4km to the east lies the sea at Sibaltan Bay (Figure 4). The landscapeis dominated by the floodplain of the Dewil river and its tributaries, covered by rice paddies,tree and vegetable crops and secondary rain forest.

In 1998 the National Museum of the Philippines began a long-term survey and excavationprogramme in the region, including Ille, other nearby caves and their surroundings (Cayron2004; de la Torre 1999; Hara & Cayron 2001; Jago-on 2000; Kress 2004; Pawlik 2004;Paz 1998; SEAICE 1999; 2000; Solheim 2000; Szabo et al. 2004; Teodosio 2004). Initialinvestigations in the east mouth of Ille Cave revealed two late prehistoric burials underlainby shell midden layers dated to c . 5000-7000 cal BP. The deepest deposits, overlying anapparent rock floor c . 1.5m below the modern ground surface, were dated to c . 10 500BP on charcoal (Szabo 2004; Szabo et al. 2004). New investigations began in 2004 uponhearing that the site had been looted and that the supposed rock floor of the east mouthtrench had been broken through by treasure hunters, exposing a further 40+cm of earlyHolocene and possibly Late Pleistocene cultural deposits (Paz & Ronquillo 2004).

319

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

Figure 2. Plan of excavations at Ille Cave 2004-2005 (E. Robles). No. 1 – east mouth excavations. The northernmost 2m-length with a bulbous end is the Szabo et al. 2004 trench; the looters’ pit dug through this and extended slightly to the east.The new excavations include both of these areas, and extend to the south and west, producing the main rectangular trench.

The Ille Cave Project was established to investigate the spatial extent, nature and antiquityof the archaeological sequences at the site (ASP 2005-6; Eusebio et al. 2006; Lewis et al.2006; Ochoa 2005; Vitales 2006). The previously excavated east mouth 1 × 2m trench wasextended by 4 × 4m (Figure 2), trowel-excavated by context and sieved (100 per cent ofcontexts). The extensive cemetery, shell midden and underlying deposits were investigatedto a depth of 2m, and the underlying sedimentary sequences exposed by the looters at thenorth end of the site to a depth of 3.6m. The main stratigraphic sequence is summarisedin Table 1. The upper parts of the east mouth sequence, including the large Neolithic-to eighteenth-century AD cemetery and mid-Holocene shell midden are not reportedhere.

Terminal Pleistocene to mid-Holocene sequence at Ille CaveThe focus of this paper is the early occupation levels identified beneath the c . 5000-7000cal BP shell midden (Table 1). These comprise a sloping 60 to 100cm-thick unit of burntdeposits in clayey silt, with intact and disturbed hearth features, flaked lithics, numerousfaunal remains and one cremated human burial (Contexts 334-807; Figure 5). The deepest

320

Res

earc

h

Helen Lewis et al.

Figure 3. View of Ille Cave, showing west (left) and east (right) mouths and platform (V. Paz).

deposits comprise steeply sloping alternating layers of very fine red and green, peagrit-sizedrounded limestone gravels in yellow-orange clay (20-30cm thick; Contexts 806, 1306, 1308)and stiff yellow clay layers (30+cm thick) (Contexts 866, 1307, 1309). Similar depositswere seen in a test pit south of the karst tower, suggesting they are not simply cave sediments,although their history has yet to be determined. The uppermost of these deposits (Context806) was devoid of cultural material. Below this, a few stone artefacts, fragments of charrednut and worked chert were recovered in association with a substantial anthropic-derivedassemblage of burnt and unburnt animal bones (Context 866; Tables 3 and 4). Twentycharcoal and bone samples from secure contexts produced a sequential range of dates fromc . 9000-11 000 cal BP, linking the well-stratified archaeological sequence to a high resolutionchronology (Table 2). A further sample of charcoal from Context 866 returned a date ofc . 14 000 cal BP. The low combustion yield and a depleted δ13C value (Tom Higham, pers.comm.) add some uncertainty to the accuracy of this date, but if confirmed it places thisdeposit well into the Pleistocene.

Deposits dated to c. 9400-11 000 cal BP

Contexts 334 to 807 comprise silty clay layers rich in fragments of decaying limestone andspeleothems, with occasional oxidised and reduced clay lumps. Limestone pebbles (someapparently burnt), charcoal, burnt and unburnt animal bones, shells and chert and obsidian

321

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

Figure 4. View of Sibaltan Bay to the south-east, showing Ille (in box) and other karst towers (V. Paz & E. Robles).

flakes occasionally occurred in discrete clusters in apparent activity areas. Several intactindividual hearths were found within these units (Table 1), along with hearth rake-outmaterial. Radiocarbon dates from these contexts span c . 9400-11 000 cal BP (Table 2).

Human foragers appear to have hunted a diverse range of local fauna from differentecological zones in the forest, including pig, deer, macaque and various small carnivores(Tables 3 and 4). The remains of turtles and snakes are also common in the archaeologicalrecord (Table 4), and even though monitor lizards are notorious scavengers and may havebeen attracted to refuse, the consistent abundant presence of their partial skeletons suggeststhat they too represent part of human prey.

Intriguingly, in the cultural sequence succeeding Context 866 deer bones diminish innumber, and by the mid-Holocene the deer is replaced by pig as the dominant hunted taxonin the assemblage. Even with the extensive sampling and sieving strategies employed at IlleCave, fish and bird bones are rare in these deposits. The few identifiable fish bones appear tobe those of freshwater Cyprinidae. Other small mammals such as bats and murids probablyrepresent inclusions in accumulating sediments, although human consumption of these cannever be wholly discounted.

The deposits contained 56 flaked stone artefacts of chert and obsidian, often occurringas isolated scatters. The raw materials have yet to be sourced, but preliminary observationssuggest that while chert may be local (Barton 2006, after Hashimoto & Sato 1973) theobsidian is likely to be imported, but is not from a known source in the Philippines (Neri2006). This is the oldest obsidian recovered from a well-dated and stratified archaeological

322

Res

earc

h

Helen Lewis et al.

Table 1. Main contexts and generalised sequence in the east mouth (east mouth excavation team).

Layers 1-2 As Szabo et al. (2004); frequent burials (not listed), cut/fill of pit 700727 Deepest inhumation; beads, conch, tusks, stone-covered BURIALS

741 Shell line dipping east732 East of 727, layer of red silt dipping east733 West of 727, layer of red silt underlying 732802 = 803 Shell layer in west section, dipping south731 & 789 Piles & layers of crushed shells, on rock fall745 Crushed shell layer746 & 747 Layers of red silt, tipping south (746) & west of rock fall (747)Rock falls (no number)771 Crushed shells under rock fall, in south end332 Shell midden (=390 = Layer 3 Szabo et al. 2004) SHELL MIDDEN

334 = 718 Organic silt under shell midden (=391 = Layer 4 Szabo et al. 2004)763 Hearth within 718 at 150cm W3N2759-760 Silt layer (759) with layer of small pebbles (760)758 Clayey silt layer; at base, cremation found on top of 336 stone336 Reddish-brown silty clay; limestone/speleothems, hearths &

burning deposits (=392 = Layer 5 Szabo et al. 2004)764 Cobbles within 336 140-160cm W2N4778 Dark brown silt; crushed shells771 Dark brown silt, frequent animal bones & turtle carapace769 ‘Melted’ limestone extending across trench (within 2004 context 336)784 As 336, but more cobble piles, surfaces and hearth remains (=337)768 Hearth/activity area; stones, frequent charred & burnt remains, chert790, 797, Burning & stone arrangements; 790 had chert flakes

801850 Similar to 336338 Clay-rich layer with frequent small limestone boulders807 Interface of 784 & 806; boulders, animal bones, shells HEARTHS

806 Compact orange clay, limestone pea grit gravels, no finds/bones866 Compact yellowish-orange clay; animal bones, antler, chert, charcoal1306 As 806, angular stones1307 As 866, but no cultural remains; angular to subangular stones1308 As 806, rounded stones1309 As 866, no cultural remains; rounded stones CLAY & GRAVELS

context in the Philippines. Undated ‘Mesolithic’ flaked obsidian has been found in Luzonin the north of the country (Scott 1968).

The lithic assemblage contains a high proportion of cores (n = 7, 12.5 per cent), allreduced using bipolar techniques with effort to control core face geometry. Several cores haveparallel longitudinal flake scars indicating the production of flakes with blade dimensions(Barton 2006). The ratio of complete flakes (n = 16) and bipolar flakes (n = 9) to cores(n = 7) is 3.6:1. The lack of early stage core reduction debitage and of other debitagefrom bipolar flaking, and the absence of hammerstones and anvils, indicate that cores werereduced elsewhere and brought to the site with other flakes. The flake assemblage comprisesmainly unretouched tools, with only two retouched flakes identified.

323

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

Figure 5. West-facing section through the east mouth main trench (T. Vitales, after V. Paz, H. Lewis & J. Carlos). The 3-Dlocation of the cremation burial in relation to this drawing is c. 2m to the west in W4N3, at around the depth of c. 140cm(top of cremation), c. 20cm below the base of the shell midden layers, which here slope down to both the south and east.

Use-wear analysis reveals that a high proportion of exhausted cores were subsequentlyused as tools. These were selected for the use of their flake margins – a unique feature of thisassemblage. Sixteen of the chert tools show use-wear in the form of edge damage or micro-traces such as micro-scarring, use polish, rounding and striations (Table 5) (Barton 2006). Asubset of six pieces demonstrates use-wear consistent with specialised functions, possibly theplaning of siliceous plants, reeds or wood. The use polish on these pieces is very distinctive,with the heaviest polish on the apex of the worked edge; this is very bright, continuous andsometimes extensively pitted. Away from the edge, polish is often reticulate and joined up,like wood-type polish, and similar to polishes described by Vaughan (1985) in experimentson tools used to plane reeds (Phragmites sp.). Most of these tools are bipolar cores with highto obtuse edge angles. These edges would be stable, robust and long-lasting. In general theused artefacts suggest that Ille was a location for the production and/or maintenance of anorganic technology, rather than a site of flaked tool production and maintenance.

324

Res

earc

h

Helen Lewis et al.

Initial archaeobotanical results show the presence of untransformed seeds, charredparenchymatous tissues, and charred wild nut fragments. The latter are of at least threetypes, one of which has been identified as probable Canarium sp., rainforest trees that aresources of edible nuts, oil and resin. This suggests the existence and exploitation of somemoist forest resources in the early to mid-Holocene.

An early cremation burial

In square W4N3 at 1.3-1.48m depth, a tight stack of fragmented burnt human bones wasfound standing on a slab of degraded fallen limestone/speleothem (Figure 6). The compactstacked organisation of the bones suggests that the cremation was originally placed withinan organic container that subsequently decayed; no feature cut was observed. The boneswere burnt, but most were not fully whitened, only charred. No charcoal was found in directassociation with the bones, which appear to have been cleaned before interment. A fragmentof calcined bone from the cremation was dated at ORAU using the method outlined byLanting et al. (2001) and yielded two dates ranging between c . 9000-9500 cal BP. To ourknowledge, these are the earliest direct dates of cremated human bones ever reported fromthe region.

The stack of bones includes at least 610 fragments from <1-8cm in size, representing c . 45skeletal elements, with no replication, suggesting that they are from one individual (Lara2006). Well-represented elements are the cranium, femora, humeri, radii, ulnae and fibulae;those with minimal representation are those with large cancellous portions (ribs, pelvis andvertebrae). The human remains underwent a series of modifications before burial, includingdisarticulation, fragmentation, burning and re-fragmentation. There are also indications thatthe cranium was skinned and the tibia defleshed. Cutmarks are found at various locations,mostly situated at or close to articular joints.

The remains probably represent a female (after Acsadi & Nemeskeri 1970; Bruzek 2002),with an age-at-death from late twenties into middle adulthood (after modern standards;Suchey in Bass 1995). A short stature (<165.2cm) is suggested based on a reconstructedleft fibula (after Trotter & Gleser 1958; Pelin & Duyar 2003). Pitting, characterised bysmall non-coalescing pores, was noted on some of the cranial bone external surfaces, andabnormal bony deposition (mixed lamellae of woven bone and sclerosis) was also observedon the anterior surface of the right femur suggesting active disease with some healing at timeof death.

DiscussionThe extent, age range and apparent intensity of the activity observed at the east mouth of IlleCave suggest long use as a place of frequent habitation during the early to mid-Holocene.Interpreted activities include hunting, food preparation, discard of food refuse, and activitiessuch as planing siliceous plant materials and wood. The lithics suggest that Ille was not aprimary stone tool production site; other types of occupation and/or tool production sitesmust therefore be located elsewhere in the landscape, and remain to be documented.

The record of diminishing deer remains at the cave reflects shifting subsistence trendsevident elsewhere on the island. Fox (1970) reported the presence of an extinct deer taxon in

325

TerminalPleistocene

tom

id-Holocene

occupationand

anearly

cremation

burial

Table 2. Radiocarbon dates from the east mouth of Ille Cave 2004-6 (compiled by H. Lewis, after T. Higham & Oxford RadiocarbonAccelerator Unit 2005-2007, pers. comm.). Results are calibrated using IntCal04 (Reimer et al. 2004). Uncalibrated dates use the half life of5568 years with isotopic fractionation corrected using the δ 13C values listed (to +−0.3 per mil relative to VPDB). The pre-treatment, targetpreparation and AMS measurement used follow Bronk Ramsey et al. (2004) & Hedges et al. (1989). Dates are presented by depth,although due to slope this does not always equate to stratigraphic sequence (e.g. [866] underlies [807] – see Table 1).

Sample Information ORAU Info Calibration

Sample Material & Context/location Sample No. δ13C Uncal BP Unmod. Cal BC/AD % Cal BP Prob.

IV-1998-P-16943 Charcoal, N1W3 100-120cm [731] OxA-14959 −23.5 5802 +− 38 4546-4767 BC 95.4% 6494-6677 0.9806687-6690 0.0036704-6716 0.017

IV-1998-P-23076 Charcoal, 100-120cm [758] OxA-16095 −29.9 5769 +− 37 4531-4715 BC 95.4% 6481-6662 1.0004580-4690 BC 60.3% 6508-6517 0.0764550-4570 BC 7.9% 6532-6574 0.394

6577-6634 0.530

IV-1998-P-23075a Crem. bone, Cremation [758] OxA-16020 −19.4 8155 +− 50 7057-7313 BC 95.4% 9006-9260 1.000(Specimen A) 7230-7250 BC 68.3% 9183-9190 0.031

7060-7190 BC 9012-9135 0.969

IV-1998-P-23075b Crem. bone, Cremation [758] OxA-15982 −25.5 8315 +− 50 7189-7516 BC 95.4% 9280-9425 1.000(Specimen B) 7331-7477 BC 68.2% 9464-9137 0.931

7189-7230 BC 0.069

IV-1998-P-15837 Charcoal, 2004r 153cm [335] L5 OxA-14898 −24.5 8545 +− 40 7531-7602 BC 95.4% 9480-9550 1.000

IV-1998-P-15839 Charcoal, 2004o 153cm [335] L5 OxA-14893 −24.3 8705 +− 40 7597-7936 BC 95.4% 9546-9784 0.9859850-9863 0.0099877-9885 0.006

IV-1998-P-20497 Charcoal, 2004 155cm OxA-16153 −27.2 8705 +− 45 7596-7938 BC 92.7% 9559-9696 1.0007890-7940 BC 2.7% 9545-9796 0.9617600-7750 BC 68.2% 9804-9817 0.011

9848-9866 0.0189875-9886 0.010

IV-1998-P-16782 Charcoal, 2004 166cm OxA-14960 −27.8 9400 +− 45 8566-8785 BC 95.4% 10515-10733 1.000

IV-1998-P-15828 Charcoal, 2004k 170cm [336] OxA-14894 −26.6 8920 +− 45 7956-8250 BC 95.4% 9905-10198 1.000

326

Research

Helen

Lewisetal.

IV-1998-P-15824 Charcoal, 2004a 170cm [336] OxA-14897 −25.6 8970 +− 45 7969-8284 BC 95.4% 9918-10071 0.41310115-10232 0.587

IV-1998-P-1583 Charcoal, 2004h 175cm [336] OxA-14899 −26.3 8799 +− 40 7715-8183 BC 95.4% 9663-9951 0.8919990-10012 0.01410024-10040 0.01010061-10132 0.085

IV-1998-P-15832 Charcoal, 2004d 177cm [336] OxA-14896 −25.8 8860 +− 45 7817-8223 BC 95.4% 9745-9748 0.0039766-10171 0.997

IV-1998-P-18309 Charcoal, 180cm [336] = [717] OxA-15817 −26.6 8795 +− 45 7650-8200 BC 95.4% 9627-9953 0.8819989-100013 0.01710022-10041 0.01210060-10134 0.08910143-10146 0.002

IV-1998-P-18310 Charcoal, 180cm [769] OxA-15766 −25.6 8830 +− 45 8020-8210 BC 95.4% 9701-9964 0.6569984-10154 0.344

IV-1998-P-15829 Charcoal, 2004g 183cm [336] OxA-14592 −25.4 9340 +− 45 8740-8470 BC 95.4% 10419-10472 0.09710476-10687 0.903

IV-1998-P-18308 Charcoal, 185cm [784] OxA-15873 −27.2 8725 +− 55 7946-7600 BC 95.4% 9549-9895 1.0007796-7610 BC 68.3% 9559-9745 0.9457815-7802 BC 9751-9764 0.055

IV-1998-P-18311 Charcoal, 185cm [784] OxA-15818 −25.4 8790 +− 40 7990-8170 BC 95.4% 9627-9939 0.9349994-10006 0.00710029-10036 0.00410064-10120 0.056

IV-1998-P-15825 Charcoal, 2004b 200cm [337] OxA-14163 −25.0 9740 +− 75 8838-9338 BC 95.4% 10786-10974 0.22010990-11032 0.24011062-11273 0.756

IV-1998-P-30189 Charcoal, N4W2 240-255cm [866] OxA-16666 −30.1 12120 +− 60 11871-12168 BC 95.4% 13820-14116 1.00068.3% 13890-14048 1.000

IV-1998-P-30194 Charcoal, N3W3 250-260cm [807] OxA-16657 −25.9 9215 +− 45 8302-8555 BC 95.4% 10252-10501 1.00068.3% 10283-10425 0.924

10467-10480 0.076

327

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

Table 3. List of hunted mammalian taxa identified in the early to mid-Holocene deposits (P. Piper &J. Ochoa). Taxonomy follows Esselstyn et al. (2004) except 1Lucchini et al. (2005). ∗Initialmorphometric analyses suggest more than one taxon represented. Analytical data to be publishedindependently.

Order Family Taxon Common name

Primates Cercopithecidae Macaca fascicularis Long-tailed macaquePholidota Manidae Manis culionensis Palawan pangolinCarnivora Felidae Prionailurus bengalensis Leopard cat

Mustelidae Amblonyx cinereus Oriental small-clawed otterMephitidae Mydaus marchei Palawan stink badgerHerpestidae Herpestes brachyurus Short-tailed mongooseViverridae Paradoxurus hermaphroditus Common palm civet

Arctictis binturong BinturongArtiodactyla Cervidae Axis/Cervus spp.∗ Deer

Suidae Sus cf. ahoenobarbus1 Palawan wild boar

Table 4. Number of individual specimens (NISP) of selected fauna recorded to date from the eastmouth, presented from youngest to oldest major mid-Holocene to late Pleistocene contexts (P. Piper& J. Ochoa).

CONTEXT Deer Pig Turtle Monitor Lizard Macaque Small Carnivores

332 (shell midden) 8 190 20 47 41 15334 0 17 3 4 7 6336 6 6 70 11 5 3769 0 10 9 5 6 1784 8 12 72 37 21 19807 16 9 43 13 13 4866 288 14 68 7 13 3

TOTAL 326 258 285 124 106 51

the archaeological record from Tabon Cave. He contrasted this with the complete absenceof deer remains in any Neolithic or Metal Age archaeological assemblages from the area, andsuggested that deer became extinct on Palawan c . 4000 years ago, but we cannot securelydate this event. Other shifts in hunting strategy have also been noted from elsewhere inthe region at around this time (e.g. Bulbeck 2003). From the zooarchaeological record atIlle Cave we know that deer were under constant, if not intensive, hunting pressure fromat least the terminal Pleistocene, and they seem to disappear from the Palawan record inthe later Holocene, at which time local foragers switched their primary hunting strategyto concentrate more heavily on pigs. The long-term decrease in deer coincides with majorchanges in environment as the proposed drier and more open vegetation of the Last GlacialMaximum was gradually replaced with moist forest into the Holocene (Bird et al. 2005;2007). The pollen record from an adjacent swamp with a basal date of c . 18 000 cal BP isbeing developed (J. Stevenson pers. comm.) as part of the Ille project to shed more light onvegetation change over the late Pleistocene to mid-Holocene for the region.

328

Res

earc

h

Helen Lewis et al.

Figure 6. Cremation burial as found (V. Paz & H. Lewis).

An apparent hiatus in cultural deposition in Context 866, currently lying between depositsdating from c . 14 000 to c .11 500 cal BP reflects broader regional trends. For instance,Semah et al. (2004; Semah & Semah 2006) note at Song Terus, Java, temporary humanabandonment at the Pleistocene-Holocene boundary, with resumption of intensive humanactivity c . 10 000 cal BP. There is obviously an interesting story to be told regionallyregarding the Pleistocene-Holocene transition; ongoing dating and research into palaeo-landform, palaeoecological and cultural change in the Ille area will further clarify landscapehistory and help develop the regional Late Pleistocene and Holocene picture.

The identification and direct dating of a human cremation burial shows that the LatePalaeolithic occupants of Ille Cave buried some of their dead where they lived and that theypractised complex burial rites, including cremation and disarticulation with interment incontainers in cave and rockshelter entrances. There are some partial parallels with other sitesin the region, including Tabon Cave where (unburnt) poorly stratified human remains werefound associated with Late Palaeolithic assemblages (Fox 1970) and Niah Cave in Borneo,where (unburnt) Late Palaeolithic inhumation burials have been identified (Harrisson1967).

329

TerminalPleistocene

tom

id-Holocene

occupationand

anearly

cremation

burial

Table 5. Summary of use-wear analysis of 16 used tools from the east mouth deep occupation sequence (H. Barton). Tool number is NationalMuseum of the Philippines accession number, beginning with IV-1998-P (site code), or an independent series (IC) where no accession number wasgiven. Individual used edges are identified by letters, where number of letters denotes number of used edges. Scars: SN (snap), F (feather), H (hinge),ST (step).

Number Type Used edge Macro Micro Rounding Striations Polish Development Mode of use Interpretation

IC001 (no #) Retouchedbrokenflake

A ST, H None Slight None Present Weak Unknown Unknown use

B H None Heavy Parallel Present Weak Cutting Cutting/slicing softmaterial – plant?

IC003 (no #) Retouchedflake

A H – Moderate None Present Weak, restricted toedge and high zonesof flake scars

Scraping Scraping medium/hardmaterial

14726 Unretouchedflake

A None SN, F, H Slight None Present Bright, patchy, smoothwith pitting andreticulate away fromthe edge

Shaving/ slicing Shaving mediumwood/plant

14727 Core A, B, C None ST Heavy Transverse Present Bright, smooth Planing Planing siliceous plant;very heavily wornedge, flattened

16310 Flake A None F None None None Weak Scraping Light duty scraping14730 Flake A None None Heavy None None N/A Scraping Scraping hard material

(residue)15228 Bipolar flake A None F None None None N/A Unknown Possibly not usedIC016 (no #) Bipolar core A, B F F Heavy Transverse Present Bright, smooth Planing Planing siliceous plant;

edge very heavilyworn, flattened inplaces

IC017 (no #) Flaked piece A, B None None Heavy Transverse Present Bright, smooth withsome pitting

Planing Planing siliceous plant;edge very heavilyworn, flattened inplaces

330

Research

Helen

Lewisetal.

Number Type Used edge Macro Micro Rounding Striations Polish Development Mode of use Interpretation

16309A Bipolar core A None None Heavy Transverse Present Bright, smooth withsome pitting

Planing siliceous plant

16309B Bipolar flake A F F None None Present Weak Shaving/slicing Medium materialB None F, SN None None Present Weak Shaving/slicing Medium material

16309C Bipolar core A None None Slight None Present Weak Planing Planing siliceous plant14775 Bipolar flake A F, H, SN, ST F Slight None Present Weak Cutting/slicing Soft to medium

materialIC044 (no #) Flaked piece A None None Slight None Present Weak Cutting/slicing Cutting/slicing soft to

medium plantIC045 (no #) Bipolar core A None None Heavy None Present Bright, reticulate Planing Planing wood or

siliceous plant14090 Bipolar

fragmentA None None Heavy Transverse Present Bright, smooth with

some pittingPlaning Planing siliceous plant

331

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

However, the interment of a cremated body in a container presages later prehistoricpractice in the region, and is significant in light of the possible longevity of this tradition. AtNiah and other cave sites, and indeed in Palawan, cremation burials in containers (mainlyurns), piles (possibly originally in baskets, wooden containers or cloth), or small pits havebeen described. These have not been directly dated, but form part of Neolithic and MetalAge cemetery contexts (Bayard 1996-7; Fox 1970; Harrisson 1967). Early, dated, cremationsfrom the larger region appear to be relatively rare. The oldest proposed cremation burialis from Lake Mungo 1, Australia, with bone fragments dated to c . 19 000 and 24 000 BP(Bowler et al. 1972; Gillespie 1997; 1998).

ConclusionStudies of early human colonisation and prehistoric adaptation to the changingenvironments of the late Pleistocene and early to mid-Holocene in Southeast Asia haveproduced evidence for complex adaptations amongst Late Palaeolithic people, as well as forsubstantial localised variations in environment over the period (Anderson 1997; Barker et al.2002; 2007; Reynolds 1993; Szabo et al. 2007). The ability of human populations to inhabitIlle Cave throughout this time, despite changing environments, reinforces the evidence fromNiah Caves (Barker et al. 2007), where it has been argued that foragers subsisted in a complexmosaic of changing environments that included open woodland, savannah and dense tropicalforest. The behavioural flexibility demonstrated by early populations in the region suggeststhat few, if any, environmental regimes restricted human dispersal (contra e.g. O’Connor2007) or subsequent success.

An early cremation burial associated with these occupation deposits shows the practice ofinterment in a container at c . 9000-9500 cal BP, much earlier than previously evident forsuch traditions in the region. In the context of this currently unique find, future efforts todirectly date Holocene human burial contexts across the region are greatly needed.

The significance for Southeast Asian archaeology of developing well-dated, stratifiedsequences with secure contextual data cannot be understated. The development of regionaland local landscape and environmental histories has been hampered through limitedapplication of high-resolution methodologies for absolute dating (see Barker et al. 2007;O’Connor 2007; Roberts et al. 2005; Szabo et al. 2007), and limited use of context-orientedexcavation strategies, both resulting in numerous uncertainties regarding artefact and sitetypologies (e.g. Lampert et al. 2003; 2004; White 2004). The archaeological sequencereported here, which also extends into the later Holocene (Szabo et al. 2004), is a rare andimportant resource for study of the ancient Philippines and Southeast Asia. Ille is, however,just one of thousands of cave sites known in the region; it is clear that we must continue toinvestigate these sites with a multidisciplinary approach strongly dedicated to understandingand dating extant stratified archaeological sequences.

AcknowledgementsThanks to the many local people, colleagues and students who worked with us in the field and laboratory, andto G. Paulino, M. & A. Laririt, E. Lin, S. Dandan, M. Monzales, the staff at Lally & Abet Cottages, M.C.Swete Kelly, K. Szabo, staff of the Oxford Radiocarbon Accelerator Unit, G. Barker, P. Bellwood, C. Gosden, C.

332

Res

earc

h

Helen Lewis et al.

French, A. Pawlik, M. Eusebio, A. Tiama, D. Jacar, A. Pagulayan, E. Dizon, W. Longacre, E. Bacus, V. Pigott,T. Reynolds, M. Spriggs, M. Carver, an anonymous referee, V. Stoukalov and especially the Solheim householdand the Paz family.

Funding: British Academy, the NERC/AHRC Orads programme, Rio Tuba & Coral Bay Nickel Mines, J.Kress, UP Diliman Archaeological Studies Program, the Solheim Foundation for Philippine Archaeology, SEAirairlines. Facilities and support: UP Diliman ASP, National Museum of the Philippines, University of OxfordInstitute of Archaeology and Research Laboratory for Archaeology & the History of Art, University CollegeDublin School of Archaeology, University of Leicester School of Archaeology & Ancient History.

Project & excavation directors HL,VP, WR & WS; co-authors excavating east mouth 2005-6 HL, VP, ML,AJC, TV, VH, HB, PP, JO, ECR, RP, WS & JS; east mouth supervisor AJC; test pit ECR & AR; geoarchaeologyHL; archaeobotany VP & AJC; human remains ML; lithics and residue analyses HB; obsidian LN; faunalremains PP and JO; radiocarbon dates HL, TH & ECR; palaeoecology JS; permits & archive WR.

ReferencesAcsadi, G. & I. Nemeskeri. 1970. History of human

life span and mortality. Budapest: Akademiai Kiado.

Anderson, D.D. 1997. Cave archaeology in SoutheastAsia. Geoarchaeology 12(6): 607-38.

ASP (Archaeological Studies Program). 2005-6. Reporton the excavation at Dewil Valley, El Nido Palawan.Unpublished report, University of the Philippines.

Barker, G., H. Barton, P. Beavitt, M. Bird, P.Daly, C. Doherty, D. Gilbertson, C. Hunt, J.Krigbaum, H. Lewis, J. Manser, S. McClaren,V. Paz, P. Piper, B. Pyatt, R. Rabett, T.Reynolds, J. Rose, G. Rushworth & M.Stephens. 2002. Prehistoric foragers and farmers inSoutheast Asia: renewed investigations at NiahCave, Sarawak. Proceedings of the Prehistoric Society68: 147-64.

Barker, G., H. Barton, M. Bird, P. Daly, I. Datan,A. Dykes, L. Farr, D. Gilbertson, B. Harrisson,C. Hunt, T. Higham, L. Kealhofer, J.Krigbaum, H. Lewis, S. McLaren, V. Paz, A.Pike, P. Piper, B. Pyatt, R. Rabett, T. Reynolds,J. Rose, G. Rushworth, M. Stephens, C.Stringer, J. Thompson & C. Turney. 2007. The‘human revolution’ in lowland tropical SoutheastAsia: the antiquity and behavior of anatomicallymodern humans at Niah Cave (Sarawak, Borneo).Journal of Human Evolution 52: 243-61.

Barton, H. 2006. A technological and functionalanalysis of Ille Cave stone tools. Unpublishedreport, Ille Cave Project, University of thePhilippines.

Bass, W.M. 1995. Human osteology: a laboratory andfield manual (4th edition). Columbia (MO):Missouri Archaeological Society.

Bayard, D.T. 1996-7. Bones of contention: the NonNok Tha burials and the chronology and context ofearly Southeast Asian bronze, in F.D. Bulbeck (ed.)Ancient Chinese and Southeast Asian Bronze Agecultures, Volume 2: 889-940. Taipei: SMCPublishing.

Bellwood, P. 1997. Prehistory of the Indo-Malaysianarchipelago (revised edition). Honolulu: Universityof Hawai’i Press.

–2005. The first farmers: the origins of agriculturalsocieties. Malden (MA): Blackwell.

Bird, M.I., I.D. Taylor & C. Hunt. 2005.Palaeoenvironments of insular Southeast Asiaduring the Last Glacial Period: a savannah corridorin Sundaland? Quaternary Science Reviews 24:2228-42.

Bird, M.I., E.M. Boobyer, C. Bryant, H.A. Lewis,V. Paz & W.E. Stephens. 2007. A long record ofenvironmental change from bat guano deposits inMakangit Cave, Palawan, Philippines. Earth andEnvironmental Science Transactions of the RoyalSociety of Edinburgh 98: 59-69.

Bowler, J.M., A.G. Thorne & H. Polach. 1972.Pleistocene man in Australia: age and significance ofthe Mungo skeleton. Nature 240: 48-50.

Bronk Ramsey, C., T.F.G. Higham & P. Leach.2004. Towards high precision AMS: progress andlimitations. Radiocarbon 46(1): 17-24.

Bruzek, J. 2002. A method for visual determination ofsex, using the hipbone. American Journal of PhysicalAnthropology 117: 157-68.

Bulbeck, F.D. 2003. Hunter-gatherer occupation ofthe Malay Peninsula from the Ice Age to the IronAge, in J. Mercader (ed.) Under the canopy: thearchaeology of tropical rain forests: 119-60. NewBrunswick; Piscataway (NJ): Rutgers UniversityPress.

Cayron, J.G. 2004. The Sinalakan turtle-shapedartefact, in V. Paz (ed.) Southeast Asian archaeology.Wilhelm G. Solheim II Festschrift: 276-91. Diliman:University of the Philippines Press.

De la Torre, A. 1999. Archaeological exploration atIlle Cave site (IV-98-P), at Barangay New Ibajay,Municipality of El Nido, Province of Palawan(January 19-February 2 1999). Unpublished report,National Museum of the Philippines.

333

Terminal Pleistocene to mid-Holocene occupation and an early cremation burial

Detroit, F., E. Dizon, F. Semah, C. Falgueres, S.Hameau, W. Ronquillo & E. Cabanis. 2002.Notes on the morphology and age of the TabonCave fossil Homo sapiens. Current Anthropology43(4): 660-6.

Dizon, E.Z. 2003. New direct dating of the humanfossils from Tabon Cave, Palawan, Philippines.Proceedings of the Society of Philippine Archaeologists1: 63-7.

Esselstyn, J.A., P. Widman & L. Heaney. 2004. Themammals of Palawan Island, Philippines.Proceedings of the Biological Society of Washington117(3): 271-302.

Eusebio, M.S., D. Arriola, J. Carlos, F. Campos, N.de la Concepcion, V. Hernandez & T. J.Vitales. 2006. The latest on the 2006 excavationseason at the Ille rockshelter. Test Pit 8: 13-17.

Fox, R. 1970.The Tabon caves. Manila: NationalMuseum of the Philippines.

Gillespie, R. 1997. Burnt and unburnt carbon: datingcharcoal and burnt bone from the Willandra Lakes,Australia. Radiocarbon 39: 225-36.

–1998. Alternative timescales: a critical review ofWillandra Lakes dating. Archaeology in Oceania 33:169-82.

Hara, Y. & J.G. Cayron. 2001. A preliminary reporton the excavation of Ille Cave, El Nido, Palawan.Hukay 3(1): 81-92.

Harrisson, B. 1967. A classification of Stone Ageburials from Niah Great Cave, Sarawak. SarawakMuseum Journal 15: 126-200.

Hashimoto, W. & T. Sato. 1973. Geologic structureof North Palawan and its bearing on the geologicalhistory of the Philippines, in T. Kobayashi & R.Toriyama (ed.) Geology and palaeontology ofSoutheast Asia 13: 145-61. Tokyo: University ofTokyo Press.

Hedges, R.E.M., I.A. Law, C.R. Bronk Ramsey &R.A. Housley. 1989. The Oxford accelerator massspectrometry facility: technical developments inroutine dating. Archaeometry 31(2): 99-113.

Heaney, L.R. 1985. Zoogeographic evidence forMiddle and Late Pleistocene land bridges to thePhilippine islands. Modern Quaternary Research inSoutheast Asia 9: 127-44.

Heaney, L.R., J.S. Walsh Jr. & T. Peterson. 2005.The roles of geological history and colonizationabilities in genetic differentiation betweenmammalian populations in the Philippinearchipelago. Journal of Biogeography 32: 229-47.

Jago-on, S.C. 2000. Archaeological program: Ille rockshelter 11 March-17 May 2000. Unpublishedreport, National Museum of the Philippines.

Kress, J.H. 2004. Ille excavations, site report. Test Pit 4:15-18.

Lampert, C., I.C. Glover, R.E.M. Hedges, C.P.Heron, T.E.G. Higham, B. Stern, R.Shoocongdej & G.B. Thompson. 2003. Datingresin coating on pottery: the Spirit Cave earlyceramic dates revised. Antiquity 77: 126-33.

Lampert, C., I.C. Glover, R.E.M. Hedges, C.P.Heron, T.E.G. Higham, B. Stern, R.Shoocongdej & G.B. Thompson. 2004.Response. Antiquity 78: 186-7.

Lanting, J.N., A.T. Aerts-Bijma & J. van derPlicht. 2001. Dating of cremated bones.Radiocarbon 43: 249-54.

Lara, M.G. 2006. An integrated report on the burialsfound in Ille Rockshelter, El Nido, northernPalawan during the 2005 excavation. Unpublishedreport, University of the Philippines.

Lewis, H.A., V. Paz, J. Kress, M.G. Lara, J.G.L.Medrana, A.J. Carlos, P. Piper, V. Hernandez,H. Barton, E. Robles, T.J. Vitales, A.Ragragio, W. Solheim II & W. Ronquillo.2006. Early occupation at Ille Cave, New Ibajay, ElNido, Palawan, Philippines: report on the 2005excavation season. Unpublished report, BritishAcademy, NERC/Orads & National Museum ofthe Philippines.

Lucchini, V., E. Meijaard, H.D. Cheong, C.P.Groves & E. Randi. 2005. New phylogeneticperspectives among species of South-east Asian wildpig (Sus sp.) based on mtDNA sequences andmorphometric data. Journal of Zoology 266: 25-35.

Merill, E.D. 1923. An enumeration of Philippineflowering plants, Volume 1 (Philippine IslandsBureau of Science Publication 18). Manila: Bureauof Printing.

Neri, L.M. 2006. Obsidian and its geochemical resultsin the Philippines. Test Pit 8: 35-7.

O’Connor, S. 2007. New evidence from East Timorcontributes to our understanding of earliest modernhuman colonisation east of the Sunda shelf.Antiquity 81: 523-35.

Ochoa, J. 2005. In dogged pursuit: a reassessment ofthe dog’s domestication and social incorporation.Hukay 8: 37-66.

Pawlik, A. 2004. A geodetic site catchment of the IlleCave platform with tilted theodolite and vectorgraphics mapping, in V. Paz (ed.) Southeast Asianarchaeology. Wilhelm G. Solheim II Festschrift:225-38. Diliman: University of the PhilippinesPress.

Paz, V. 1998. Archaeological survey of El Nido,Palawan, Philippines. Unpublished report, NationalMuseum of the Philippines.

Paz, V. & W. Ronquillo. 2004. Report on thePalawan Island Palaeohistoric Research Season for2004. Unpublished report, University of thePhilippines.

334

Res

earc

h

Helen Lewis et al.

Pelin, I.C. & I. Duyar. 2003. Estimating stature fromtibia length: a comparison of methods. Journal ofForensic Science 48(4): 708-12.

Reimer, P.J., M.G.L. Baillie, E. Bard, A. Bayliss,J.W. Beck, C. Bertrand, P.G. Blackwell, C.E.Buck, G. Burr, K.B. Cutler, P.E. Damon, R.L.Edwards, R.G. Fairbanks, M. Friedrich, T.P.Guilderson, K.A. Hughen, B. Kromer, F.G.McCormac, S. Manning, C. Bronk Ramsey,R.W. Reimer, S. Remmele, J.R. Southon, M.Stuiver, S. Talamo, F.W. Taylor, J. van derPlicht & C.E. Weyhenmeyer. 2004. Intcal4terrestrial radiocarbon age calibration 26-0 ka.Radiocarbon 46:1029-58.

Reis, K.R. & A.M. Garong. 2001. Late Quaternaryterrestrial vertebrates from Palawan Island,Philippines. Palaeogeography, Palaeoclimatology,Palaeoecology 171: 409-21.

Reynolds, T.E G. 1993. Problems in the stone age ofSouth-East Asia. Proceedings of the Prehistoric Society59: 1-15.

Roberts, R.G., M.J. Morwood & K.E. Westaway.2005. Illuminating Southeast Asian prehistory: newarchaeological and palaeoanthropological frontiersfor luminescence dating. Asian Perspectives 44(2):293-319.

Scott, W.H. 1968. A critical study of the Prehispanicsource materials for the study of Philippine history.Manila: University of Santo Tomas Press.

SEAICE (Southeast Asian Institute for Culture &Environment). 1999. Archaeological study: El Nidocommunity based conservation and developmentprogram. Unpublished report, Philippine RuralReconstruction Movement.

SEAICE (Southeast Asian Institute for Culture &Environment). 2000. A brief summary of thearchaeological exploration of Ille Cave and rockshelter. Unpublished report, University of thePhilippines.

Semah, F. & A-M. Semah. 2006. Palaeolithicsettlements in the Southeast Asian archipelagos: anIndonesian perspective, in T. Simanjuntak, M.Hisyam, B. Prasetyo & T.S. Nastiti (ed.)Archaeology: Indonesian perspective: 148-61. Jakarta:Indonesian Institute of Sciences InternationalCenter for Prehistoric & Austronesian Studies.

Semah, F., A.-M. Semah, C. Falgueres, F. Detroit,T. Simanjuntak, A.-M. Moigne, X. Gallet & S.Hameau. 2004. The significance of the Punungkarstic area (Eastern Java) for the chronology of theJavanese Palaeolithic, with special reference to theSong Terus cave. Modern Quaternary Research inSoutheast Asia 19: 45-61.

Solheim, W.G. II. 2000. Excavations of Ille rockshelter, Palawan, Philippines (Preliminary report).Unpublished report, National Museum of thePhilippines.

–2006. Archaeology and culture in Southeast Asia:unravelling the Nusantao. Manila: University of thePhilippines Press.

Szabo, K. 2004. Technique and practice: shell-workingin the Western Pacific and Island Southeast Asia.Unpublished PhD dissertation, Australian NationalUniversity.

Szabo, K., M.C. Swete Kelly & A. Penalosa. 2004.Preliminary results from excavations in the easternmouth of Ille Cave, northern Palawan, in V. Paz(ed.) Southeast Asian archaeology. Wilhelm G.Solheim II Festschrift: 209-24. Diliman: Universityof the Philippines Press.

Szabo, K., A. Brumm & P. Bellwood. 2007. Shellartefact production at 32,000-28,000 BP in IslandSoutheast Asia. Current Anthropology 48(5): 701-23.

Teodosio, S.F.R. 2004. Preliminary report of theMakangit Cave investigation, in V. Paz & W.Ronquillo (ed.) Report on the Palawan IslandPalaeohistoric Research Project for 2004.Unpublished report, University of the Philippines.

Trotter, M. & G.C. Gleser. 1958. A re-evaluation ofestimation of stature based on measurements ofstature taken during life and of long bones afterdeath. American Journal of Physical Anthropology 16:79–123.

Vaughan, P.C. 1985. Use-wear analysis of flaked stonetools. Tucson: University of Arizona Press.

Vitales, T.J. 2006. Pagsilip sa Isang NatatangingArtefak ng Ille (Looking at a special artefact in Ille).Test Pit 8: 29-30.

White, J.C. 2004. Comment on dates from aresin-coated sherd from Spirit Cave, Thailand.Antiquity 78: 184-6.

335