ChapterPDF Available

Rocks, pottery and bird bones: new evidence on the material culture of Isle of Pines during its 3000-year-long chronology

Isle of Pines (or “Ile des Pins” in French) is
a small island located south of mainland
New Caledonia (Fig. 1) and is famous for its
white sand beaches, striking blue waters and
beautiful settings. It has also, over the years,
been the focus of some archaeological work,
thanks to the discovery in 1948 of Lapita
site KVO003 of St-Maurice/Vatcha. is
site, one of the most famous Lapita sites in
New Caledonia, is still frequently included
in scientic publications, even though the
last excavations were undertaken at this
location in 1995-1996 (Sand 1999). e island
is considered to be a kind of “archaeological
nightmare” due to the presence of no less
than 200 earth mounds (interpreted as
tumuli”) upon the islands red-earth plateau.
ese curious and mysterious structures also
exist on mainland New Caledonia, but are
present in much fewer numbers than on the
Isle of Pines. Moreover the number present
upon the plateau of the island may even be
outnumbered by those on the surrounding
forest plains.
Our 4 year research project on Isle of Pines
(Lagarde 2012) consisted of an inventory
of surface sites in the island’s dierent
environments, as well as carrying out a small
number of test-pit excavations, in order to
obtain a clearer picture of the past. One of
these test-pits was excavated in a rock shelter
(with the code name KTT006) situated in the
middle of the present rainforest area, on the
east coast of the island. is paper presents
the general results from the analysis of
material culture obtained from this dig, thus
providing new insights into the dynamics that
occurred upon the island during its 3000 year
history. Aer a general introduction on the
environment of Isle of Pines, the stratigraphy
of the excavated test-pit and a detailed list of
anthropic artefact types will be presented,
followed by a nal discussion on social
relationships and exchange mechanisms from
a chronological point of view.
Louis Lagarde and André-John Ouetcho
Rocks, pottery and bird bones:
New evidence on the material culture of Isle of Pines
(New Caledonia) during its 3000-year-long chronology
General setting
Isle of Pines is the southernmost inhabited
island within the Melanesian crescent,
located between 167° 25’ and 167° 35’E and
22° 32’ and 22° 42’S, it has a globally elliptical
shape and is 18 km long and 14 km wide. It
has a 60 km perimeter and a 135 km2 global
surface. Many islands lie around it which,
when combined with Isle of Pines, have a
total global surface of 158 km2 (Pisier 1972).
e natural environment can be divided in
two main types (Fig. 2):
1. a at plateau, 80 to 100 m high in the central-
west area of the island, continued in the
south by a rather small mountainous range,
which climbs to a height of 262 meters at
N’ga Peak. is geological environment is
linked to the southern mainland of New
Caledonia, and is a part of the Norfolk ridge
(Maurizot 2003: 10). Soils are derived from
the degradation of ultrabasic rocks that
come from a very deep layer of the earth’s
crust that nowadays covers Grande Terre,
Isle of Pines and the Bélep archipelago due
to a subduction process which ended in the
Eocene period, 37 million years ago;
2. a more recently elevated fossil coral fringing
reef, which is now a low plain only a few
meters above sea level, circling the island.
e origin and date of the reef upliing
remains unclear, as it is similar to the one
Figure 1. General map of the New Caledonian archipelago.
Rocks, pottery and bird bones
Figure 2. General geological environment of Isle of Pines (GEOREP, New Caledonia GIS).
Louis Lagarde and André-John Ouetcho
Figure 3. Map of rockshelter KTT006 and location of excavation. (C. Sand).
Rocks, pottery and bird bones
that caused the creation of the Loyalty
Islands, but is also similar to the elevated
reefs of the Yaté zone in southern Grande
Terre. Either way, both of these phenomena
occurred between the late Miocene and the
Holocene (Maurizot 2003: 14).
e soils generated by this environment
consist of brown rendzines that are generally
thick (Cherrier 1986: 38), with important
amounts of calcareous rock inducing a rather
high pH (±8). ese pedological characteristics
have allowed the creation of rich tropical
forest, with species similar to those present
on mainland New Caledonia and the Loyalty
Rockshelter KTT0 06 and excavation
In this forest zone of calcareous background,
a short survey was undertaken in March 2009
in order to nd a suitable rock shelter for our
excavation. e survey particularly focused
on the Truete tribe’s zone, for which we have
ethnographic data mentioning numerous
funerary deposits (Guiart 1963: 216-222,
Lambert 1900: 274-288). is area possesses
an uplied coral cli, which is up to 55m
high and orientated along a N-S axis, oering
numerous small caves, many of which contain
evidence of funerary practices. One of the
archaeological sites uncovered, KTT006, had
a number of characteristics that pointed to
a long period of use. Its main characteristic
is anthropogenic ashy soil that is more than
90 cm deep in some spots. is shelter faces
the ancient horticultural swamp zone of site
KTT003. Furthermore, it features a large
conical midden, that is located in the outside,
unprotected area of the cave connected with
ancient garbage disposal. ese factors led to
the excavation of a 3.5 m2 test-pit in September
2009 (Fig. 3).
Placed on a N-S axis, this test pit was
undertaken in a well-protected area, in
between a very large fallen calcareous boulder
on the south and the actual natural cave wall
on the north side. From North to South,
the smaller pit was named D5, while the
1m2 pits were named D6, D7 and nally the
southernmost one D8.
Four funerary structures were ultimately
found during the dig, two being secondary
deposits of random human bones and two
primary interments, one of a young child
(6 months - one year old) in squares D5-D6
and the other of an adult in square D7. Both
Figure 4. Radiocarbon dates obtained on charcoal and bone samples from site KTT006.
Louis Lagarde and André-John Ouetcho
of them remained mostly untouched by
our excavation and none of the four will be
discussed in this paper. As a general remark,
it can be stated that the stratigraphy of this
part of the rock shelter was not that heavily
disturbed by later activities or natural growth
of roots. Furthermore, between the anthropic
layers there was no sterile layer that would
have indicated a long gap in the use of the rock
shelter itself. Both observations indicate that
this site oers a wide chronological span of
the island’s history, as can be demonstrated by
the 14 radiocarbon dates obtained on charcoal
or bone samples (Fig. 4).
e general stratigraphy can be simplied
by connecting archaeological layers to
chronological data, thus dening 6 main
periods in the rock shelter’s occupational
history (Fig. 5), starting from deep layers
between 2800 BP and 2000 BP, and an
important cultural layer dating to around
1900 BP. e rst millennium AD can be split
between the period from 1900 to 1400 BP, and
1400 to 1000 BP. e second millennium AD
encompasses the pre-European layer of 1000-
200 BP, and nally the layers that compose the
rock shelter’s modern oor surface.
Archaeological results
Ceramic artefacts uncovered in rock shelter
KTT006 broadly encompass the main
characteristics and types as dened for
southern Grande Terre, with two noteworthy
exceptions: Lapita dentate-stamped pottery
and Puen pottery. e absence of the former
can be explained by the fact that the earliest
date obtained in the rock shelter comes from
a small natural pit in the ground lled with
Figure 5. 6 main cultural layers at site KTT006.
Rocks, pottery and bird bones
charcoal and tiny unidentiable
ceramic fragments. Furthermore,
the forest environment could
have supplied old wood for the
rock shelter’s inhabitants, causing
radiocarbon dates (especially
2810±40 BP, calibrated 1050-
890 BC, 2 σ, Beta 284698) to be
slightly older than the actual
human occupation. e oldest
clearly-identiable ceramics are
those visible on the natural ground
surface, but rmly embedded in
the rst main ashy anthropic layer,
which leads us to believe that they
might be slightly older than the
dates of 2500±40 BP (calibrated
790-490 BC, 2 σ, Beta-286970)
and 2450 BP (calibrated 760-400
BC, 2 σ, Beta-283044) that we
obtained. Absence of the Puen
pottery type, dating to the second
half of the rst millennium BC
in the southern half of mainland
New Caledonia, is in keeping with
the results of the intensive surface
survey undertaken since 2006 on
Isle of Pines, where no potsherd
corresponding to this particular
ceramic type could be found. is
indicates that internal relations
within the archipelago broke down
at this time.
Typologically, ceramic artefacts
related to the rst millennium BC
can be divided into three categories:
1. rather thick potsherds (0.7 cm
to 0.9 cm in width), displaying
a dark brown, very fragile clay
with evidence of very ne coral
sand temper. is pottery type
has relatively high amounts of
Figure 6. Le: late Lapita sherds from KTT006. Right: similar motif on a
potsherd from Loyalty island Lapita site LPO023 (Maré) (Sand 2010: 132).
Louis Lagarde and André-John Ouetcho
motifs but the paddle-impressions appear
quite crude, perhaps indicating a transitional
period in which potters had not yet understood
the full decorative possibilities oered by the
paddle itself.
3. all of the remaining sherds from the rst
millennium BC correspond to a thin (0.3 to
0.6 cm) paddle-impressed pottery type known
on mainland New Caledonia as the Podtanéan
tradition, with out-turned lips and carinated
vessel forms (Fig. 7). Most paddle impressions
are either highly regular or smoothed out,
the few low-quality crude examples appear
connected with Plum pottery dating to the
rst century AD (as in square D8, unit #21,
55-60 cm).
e Puen tradition is known to have
replaced the paddle-impressed Podtanéan
tradition in the south of Grande Terre during
the last few centuries BC (Sand, Bolé, and
Ouetcho 2011). Here, on Isle of Pines and
specically at KTT006, the Podtanéan pottery
type seems to have been in use until the rst
Figure 7. Paddle-impressed potsherds from KTT006.
decoration (4 out 7 sherds are decorated) and
has evidence indicating the presence of zigzag
decoration. e motifs present were impressed
with a non-dentate and possibly curvilinear
tool. is ceramic type has been found in
three other surface sites on Isle of Pines and
is interpreted as a late-Lapita transitional
type, between the dentate zigzag Lapita sherds
—as found on the Loyalty Islands, central
Vanuatu(Sand and Bedford 2010: 278), in
addition to other sites in Oceania— and the
following type (Fig. 6).
2. three ne potsherds (0.4 cm in width) show
paddle-impressions with a superimposed
zigzag motif. e temper in these sherds is
also ne. e other main characteristic is
their solidity aer almost three millennia
compared to those of the rst type. e zigzag
decoration here seems to be incised rather than
impressed, showing evidence of similarity
with the incised paddle-impressed pottery
types discovered on the Loyalty Islands (Fig. 7,
bottom example). ese sherds bear complex
Rocks, pottery and bird bones
century AD (1910±40 BP, calibrated 10-210 AD,
2 σ, Beta-283041), appearing in archaeological
layers along with sherds of the Plum tradition.
is shows a strong evolutional similarity with
the Loyalty archipelago, where the Podtanéan
tradition appears to be the last ceramic type
to be imported from Grande Terre during the
period beginning in the rst millennium AD.
e Plum pottery type is characterized
by thicker pots with tempers of larger size
and the use of sedimentary clay rather than
mangrove or estuary clays. Its other unique
characteristics are morphological and
technological: a) the use of handles, unique
in the islands ceramic history (Fig. 8) and b)
the use of clay coils rather than slabs in the
construction process. Present throughout
the whole rst millennium AD, the most
recent stratigraphic unit where it was found at
KTT006 is #12, lying immediately under #11,
the bottom of which was dated to 970±40 BP
(calibrated 1000-1160 AD, 2 σ, Beta-286968).
Introduction of this pottery type, which
characterizes the material culture of southern
mainland New Caledonia during this period,
shows a renewal of relations with that area, as
well as a ceramic evolution that, for the rst
time, deviates from the pattern of ceramic
use identied in the Loyalties and central/
southern Vanuatu. Due to the isolation of the
Loyalty group from Grande Terre during the
rst millennium AD and the development of
diering indigenous ceramic traditions within
the Vanuatu archipelago, the Plum pottery
tradition was never present in these areas.
e only known ceramic type dating
to the second millennium AD in southern
Grande Terre is Nera pottery, which is
present in unit #11 of site KTT006 (Fig. 9)
and dates to 970±40 BP (calibrated 1000-
1160 AD, 2 σ, Beta-286968). e presence
of Nera pottery in combination with Plum
pottery at KTT006 indicates that there was
continuous interaction going on between
the Isle of Pines and southern mainland
New Caledonia over the last two millennia.
is pottery type can be divided into two
main morphological categories. However
in both cases technological criteria such as
construction technique, clay type, tempers
and overall shape remain the same. e rst
category includes pots that are generally
spherical in shape, with vertical rims and
incised decoration present towards the lip.
e second category includes pots that have a
atter globular shape and are associated with
pustule motifs, a decorative technique that
oen weakens the vessel, which are generally
located near the widest part of the pot. No
evidence has been identied to suggest that
Figure 8. Partial reconstruction of a vessel from the Plum tradition, KTT006.
Louis Lagarde and André-John Ouetcho
these two categories of pottery represented
the evolution of the Nera pottery tradition
through time. erefore we believe that they
were contemporary and thus may have been
used for separate uses or functions
Lithic akes
e lithic collection uncovered from rock
shelter KTT006 consists almost entirely of
chert debris or small akes. e most striking
aspect of the collection is that it is restricted to
two phases of the island’s history, specically,
the rst millennium BC and the recent layers
dating to the second millennium AD. e
collection associated with the former consisted
of a total of 37 chert fragments of minimal size
(≤1cm2), while the latter consisted of a series of
187 chert fragments. A number of particularly
Figure 9. Decorated and undecorated rims of the Néra tradition, KTT006.
Rocks, pottery and bird bones
well-dened akes were unearthed in
this series, consisting of triangular
discoid akes resembling pseudo-
Levallois points (Fig. 10). Without
the presence of cores in the KTT006
assemblage, it is not possible to state
if the aking sequence corresponds
to the one determined for chert akes
of the Lapita period (Forestier 1994,
Forestier 1996a, Forestier 1996b,
Forestier 1999, Lagarde 2004, Sand
2010), however the nal products do
possess a striking resemblance.
Disappearance of aked material
during the rst millennium AD leads
us to believe that even though the
presence of Plum pottery indicates
interaction was occurring between
Isle of Pines and southern mainland
New Caledonia, the exchange of
akeable material was still rare
during this period. Although it could
correspond to a change in the use of
the rock shelter at this period in time,
it is more likely that pottery exchanges
or the exchange of technological
innovations via cultural interactions
was undertaken with groups within
a very limited area, such as the Yaté-
Goro zone on the southern tip of
Grande Terre. Siliceous rocks are not
present in that part of southern New
Caledonia, nor on Isle of Pines, and
therefore they had to be imported
from further north at that time.
While they were probably a very
sought-aer resource back then, the
scarcity of siliceous artefacts from the
rst millennium AD in rock shelter
KTT006 accounts for a change in the
type of importation networks. It is
likely then, that rather than the large-
scale exchanges that occurred within
Figure 10. Discoid akes from site KTT006.
Louis Lagarde and André-John Ouetcho
in the Isle of Pines’ geological environment.
e two types of beads can be described as:
1. micro beads (≤5 mm wide, ≤2 mm thick)
which are perfectly-rounded pierced steatite
discs. eir nal use is unknown, even if they
resemble the Conus shell at beads found in
rock shelter KTT006 and many other sites of the
New Caledonian archipelago (Sand 2001). ey
seem to have been part of larger ornaments,
possibly similar to the armbands known in
Figure 11. Steatite beads from site KTT006.
the wider networks of the rst millennium
BC and the second millennium AD, exchange
with Grande Terre during this period occurred
on a smaller scale.
Lithic beads
One of the most intriguing aspects of sorting
out the rock shelter’s sieving residues was the
discovery of two distinct types of green or
black steatite beads (Fig. 11). is so ultrabasic
rock, sometimes called soapstone, is not found
Rocks, pottery and bird bones
Vanuatu (Bonnemaison et al. 1996: 199). ese
beads are quite ancient: although they are not
present in the older layers, they appear alongside
Plum pottery in the early stratigraphic units of
the rst millennium AD (square D6, 65-70 cm
and square D8, 55-60 cm). ey are also found
during the second millennium AD, indicating
that the ancient exchange mechanisms that
facilitated the movement of steatite between
Isle of Pines and Grande Terre were maintained
for almost 2000 years.
2. large thick beads (≥1 cm wide and
thick), of diverse shapes (oval to circular),
resembling the nephrite or serpentine beads
found on traditional Kanak necklaces that
are specically worn by women (Leenhardt
1937: 114-115). ese beads are mostly found
in the stratigraphic units dating to the last
millennium (only one is found in square D6,
40-45 cm), indicating production occurred
during the last 1400 years.
Finally, several pieces of raw steatite show
evidence of sawing, which implies that raw
material was imported rather than fully
nished beads.
Bone artefacts
Several bone needles were uncovered from
site KTT006. Made of seabird radius wing-
bones, they exhibit a point, a small hole on the
other end, and a shiny patina connected with
intensive use (Fig. 12). One side of the bone
was to be pierced, allowing the thread to come
through the natural section of the radius. e
production sequence, requiring the drilling
of a small hole in the surface of a bird bone
accounts for a high degree of specialization.
eir use remains unknown, although the
construction of traditional Kanak slingshot
stone bags via ne knitting has been known
to occur in recent times. Additionally Kanak
traditional currency is usually carried in a
barkcloth or fabric envelope closed by the use
of a similar bone needle. On mainland New
Caledonia, these needles appear to be carved
out of human bone or the corresponding totem
animal (Leenhardt 1930: 327, Patouillet 1873:
219) and are typologically quite dierent. In
archaeological surveys some smaller bone
needles were found, for example on Ouvea
(site LUV030), but they are shorter (possibly
broken) and are not perforated in the same
fashion as those at KTT006. e only place
within the New Caledonian archipelago
bearing large numbers of intact bone needles
similar to those found in KTT006 is Walpole
Island, an isolated, now uninhabited uplied
coral cli located 180 km away from Grande
Terre. e island’s extreme environment
and particular geology led to a complex and
diversied tool industry using giant clam shell
and bird bones as raw materials (Sand 2002).
e smaller examples of bone needles collected
on Walpole Island show striking similarities
with those unearthed at site KTT006. e
recurring presence of Walpole in Isle of
Pines’ oral traditions leads us to think that
the needles were actually obtained through
exchange mechanisms. Oral traditions from
Maré Island in the Loyalty group mention
expeditions to Walpole to collect seabird
feathers (Dubois 1975), showing that this
distant island was included in voyaging routes
in ancient times. Chronologically they have
been uncovered in several layers, including the
early centuries of the rst millennium AD (one
in unit #21, dated 1910±40 BP (Cal AD 10-20,
2 σ, Beta-283041). Most of the bone needles
were found in layers dating to 1700±40 BP
(unit #16, Cal AD 240-420, 2 σ, Beta-286969).
However, they have also been found in more
recent layers; the latest of which is found in
square D5, 15-20cm (unit #8), a layer dated to
Louis Lagarde and André-John Ouetcho
Figure 12. Seabird bone needles, KTT006.
between 970±40 BP (Cal AD 1000-1160, 2 σ,
Beta-286968) and 510±40 BP (Cal AD 1330-
1340 et AD 1400-1450, 2 σ, Beta-283039). is
potentially indicates an ongoing exchange
relationship between Isle of Pines and Walpole
that lasted well over a millennium.
Faunal remains
From the faunal remains excavated from
KTT006, three bird bones were interpreted
as chicken. One specically, the metatarsus
of a male individual, bears a spur indicative
of the species Gallus gallus (Linnaeus 1758).
Rocks, pottery and bird bones
Its presence in a pre-European layer (square
D6, unit #11, 15 cm) is particularly intriguing.
Direct radiocarbon dating of the bone gave
the date of 560-630 BP, coherent with dates
obtained from charcoal samples in nearby
stratigraphic units.
No other New Caledonian archaeological
site has yet shown evidence of pre-European
introduction of chicken. However, evidence
of ancient Gallidae does exist in some oral
traditions, even if confusion exists as to what
extinct species these traditions are referring to.
Contacts between Western Polynesia and New
Caledonia are well documented within oral
traditions, particularly on Isle of Pines where
Tonga is mentioned in several illustrious
stories. Yet, archaeologically, evidence of
these contacts has remained dicult to nd.
Introduction of the jungle fowl Gallus gallus
centuries before its European subspecies
G. gallus ssp. domesticus, could be a rst step
to a better denition of recent long-distance
contacts between Southern Melanesia and
Western Polynesia.
Discussion: e point of view of
Isle of Pines in the archipelago’s
e Lapita period
Long distance exchange between New
Caledonia and other parts of the Pacic,
and local interactions within the New
Figure 13. Long-distance exchanges of goods during the Lapita cultural complex in New Caledonia.
Louis Lagarde and André-John Ouetcho
Caledonian archipelago, are well documented
for the Lapita period. A solid repertoire of
local Lapita motifs, in combination with
data upon pottery exchange derived via
dierential temper analysis indicate that the
Southern Melanesian Lapita complex was
a strongly structured cultural entity (Sand
2010). Furthermore, recent developments in
the analysis of lithic material has shown that
mainland New Caledonia was a constant
supplier of akeable siliceous rocks to areas
with poor local environments (e.g. Loyalty
islands, Isle of Pines) during the Lapita period
(Lagarde 2004, Sand 2010). Excavations on
Isle of Pines, and particularly at KVO003,
have anchored the island within the Southern
Melanesian Lapita system. On a broader scale,
the presence of Talasea obsidian in the Lapita
layers of various local sites indicates relations
with regions as far west as New Britain (Sand
and Sheppard 2000), and thus provides a link
between New Caledonia (and particularly Isle
of Pines) with the rest of the Lapita cultural
complex (Fig. 13).
e end of the rst millennium BC
As time went by local evolutions in traditional
material culture within the archipelago
took place, and by the second half of the
rst millennium BC several major changes
can be noticed. On most areas of mainland
New Caledonia paddle-impressed pottery
of the Podtanéan tradition was gradually
being replaced, particularly in the south,
with a new type of globular pot known as
Puen pottery. However, on Isle of Pines
and Loyalty Islands, perhaps owing to
somewhat dierent reasons in each case,
this particular cultural innovation was
absent. In both areas the Podtanéan ceramic
types continued to be used well into the rst
millennium AD. Similarities in the evolution
of material culture in the remote islands
within the New Caledonian archipelago
must therefore be examined. e lack of
Puen pottery demonstrates that there was
a break in short distance relations between
southern Grande Terre and these islands at
this time. It explicitly demonstrates a break
in short distance relations at this time with
southern Grande Terre. Siliceous rocks were
probably exchanged on a wider scale and on
dierent routes within the archipelago than
pottery-making techniques, and thus they
continued to be imported into Isle of Pines
up to the beginning of the rst millennium
AD (Fig. 14).
e rst millennium AD
e continual increase in population in a
malaria-free environment over a millennium
and the use of extensive burning for
cultivation practices seem to have led to the
rst widespread large-scale tensions within
the archipelago, consistent with dramatic
changes in the traditional exchange patterns
and social practices.
On the Loyalty Islands, where pottery
cannot be produced because of a lack of
quality clay, imports from Grande Terre
seem to have stopped during this period. On
Isle of Pines, the presence of Plum pottery
indicates that links with southern mainland
New Caledonia were re-established. e
majority of the ceramic production in this
area is potentially indigenous, as it is known
to have high quality clay sources; therefore,
the introduction of the Plum tradition reveals
not only that pottery exchanges occurred, but
also more importantly, that the transmission
of production techniques linked to that
particularly innovative pottery type also
Rocks, pottery and bird bones
Figure 14. Isle of Pines within the New Caledonian archipelago at the end of the rst millennium BC.
Imported lithic material during the rst
millennium AD is very scarce, indicative of
profound changes occurring with regards
to the exchange mechanisms between
Grande Terre and Isle of Pines. Siliceous
rocks could have come from sites located in
close proximity to the area, such as the Yaté-
Goro-Unia zone of southern New Caledonia,
which also had to import such resources,
thus explaining the minimal number of lithic
artefacts in the corresponding stratigraphic
units of site KTT006.
Exchange networks were not abolished
altogether with Grande Terre during this
period, but they were functioning in a
dierent manner. For example, steatite, used
in the production of small discoid beads
and sourced from non-lateritic zones of
mainland New Caledonia, is being imported
for the rst time. is represents the creation
of new exchange relationships within the
Furthermore, the presence of t ypologic ally-
unique seabird bone needles from Walpole
Island, on Isle of Pines in contexts dated to
as early as the rst centuries AD, suggests
that long-distance travel was occurring
between these two areas from an early date.
Evidence of similar needles has been found
in other sites of Grande Terre and the Loyalty
Islands, yet their preservation state and exact
typological characteristics exclude them as
being from the same operatory chain as the
ones made on Walpole and the ones used on
Isle of Pines (Fig. 15).
Louis Lagarde and André-John Ouetcho
Figure 15. Isle of Pines within the New Caledonian archipelago during the rst millennium AD.
e second millennium AD
It is generally accepted that the unique social,
cultural, linguistic, and political situation
present within New Caledonia during the
rst millennium AD generated the ‘Cultural
Traditional Kanak Ensemble’, thus giving
birth to the Kanak traditional societies
encountered by European navigators and
missionaries in the late 18th and early 19th
On Isle of Pines, most Kanak traditional
cultural markers have been uncovered
archaeologically, whether during survey
campaigns, whereby large horticultural
terraces and yam elds were identied, or
within the site of KTT006. In the rock shelter,
the main cultural markers are ceramics of the
Néra tradition. e presence of Néra tradition
pottery at the rock shelter and the associated
pottery production practices that go along
with it, show repeated contacts with southern
Grande Terre (similar examples of sherds can
be seen in Sand and Ouetcho 1992). In the
same way, small pierced shells and carved
oyster shell pendants uncovered in superior
layers seem connected with the use of Kanak
traditional money. is is also suggested by the
presence of Trochus shell armbands, which are
consistent with assemblages derived from New
Caledonia during this period. Additionally,
the reappearance of siliceous rocks in the
rock shelter indicates the presence of new
long-distance exchanges with mainland
New Caledonia. Finally, a very small piece
of distinctive Ouen Island “jade”, actually a
semi-nephrite, was also uncovered; illustrating
the well-known place of Isle of Pines within
Rocks, pottery and bird bones
the important “jade cycle” described by
anthropologists in the rst half of the 20th
centur y.
Yet, several distinctive particularities of the
KTT006 assemblage give us a more precise
image of Isle of Pines’ external relations at
this time. Firstly, steatite bead production,
and thus the importation of raw materials
that started at the beginning of the rst
millennium AD, continues during the second
millennium AD, showing that at least some
cultural exchange mechanisms are already
in place when the traditional Kanak societies
begin to appear. Secondly, some circuits,
not necessarily connected with traditional
Kanak exchange routes, are also maintained
throughout the second millennium AD, likely
due to the geographical position of the Isle of
Pines. For example, bone needles seemingly
from Walpole Island continue to be imported
into the Isle of Pines.
Oral traditions mention the introduction
of new horticultural species from Vanuatu
through the Loyalty Island group. While it
is dicult to account for any specic dietary
changes without further study, contacts with
southern Vanuatu and its islands, under
Western Polynesian inuence, seem highly
probable. Evidence of chicken bones in recent,
but still pre-European, layers at KTT006
prove that the introduction of the jungle
fowl within the New Caledonian archipelago
occurred at least several generations before the
introduction of its domesticated counterpart,
Figure 16. Isle of Pines within the New Caledonian archipelago in the second millennium AD.
Louis Lagarde and André-John Ouetcho
Gallus gallus domesticus. Isle of Pines oral
traditions mention Tonga and “Kiamo
(Aneytium or Tanna), so it is a complex task
to determine whether these jungle fowl were
brought directly from Western Polynesia or
through Southern Vanuatu. Yet they complete
the picture of a complex, and intense, set of
distant relations and contacts during the few
centuries prior to European contact (Fig. 16).
e excavation conducted on Isle of Pins, at
rockshelter KTT006, oers new insights on
the existence and consistency of exchange
mechanisms within the New Caledonian
archipelago. Beginning with the Lapita
period, where New Caledonia was probably
fully interdependent and interrelated, local
evolutions over time have produced a variety
of localized changes in social relations and
exchange mechanisms. Arrival of new people
and contacts with more distant archipelagoes
eventually led to the introduction of new
species of commensals such as chicken, and
new species of tubers for daily consumption.
While imports and the introduction of
technical innovations to Isle of Pines have been
discussed, exports have yet to be documented.
Further excavations on New Caledonian sites
may one day, provide evidence of two-way
interactions. e analysis of clay associated
with Plum pot sherds could display the
exchange of pots with southern Grande Terre
during the rst millennium AD. Similar
testing could be done upon Podtanéan and
Néra potsherds to see if pottery made on Isle
of Pines made its way to the Loyalty Islands,
either during the rst millennium BC or
the second millennium AD. Furthermore,
the existence of a few tumuli in southern
Grande Terre could been seen to indicate the
introduction of funerary practices from Isle
of Pines, as at least some of these mounds
were used as burial mounds, between 2200
BP and 1500 BP.
is paper could not have been produced
without the constant help of the IANCP and
its director, C. Sand. e four-year mission
on Isle of Pines is a PhD project that, without
his help, nor the help of the entire IANCP’s
sta, could not have been completed.
Dr Scarlett Chiu, of Academia Sinica,
easily accepted an oral presentation on this
topic at the Cross-regional Comparison of
Ancient migration and Exchange Patterns
Conference, Academia Sinica, Taipei, Taiwan,
September 2012, and should be acknowledged
as well.
e people of Isle of Pines, who have
walked, worked with us and helped us daily
in so many ways over the years, should
also be cited. ey are too numerous to be
presented personally, but we are sure they will
nd interest in understanding their past in
a dierent way than through the sole aspect
of oral history.
References Cited
Bonnemaison, J., K. Human, C. Kaufmann, and
D. Tryon. Editors. 1996. Vanuatu, Océanie: Arts
des îles de Cendre et de Corail. Paris: Réunion des
Musées Nationaux.
Cherrier, J.-F. 1986. La forêt de l’Ile des Pins. Bois et
Forêts des Tropiques 211:35-46.
Dubois, M.-J. 1975. Mythes et Traditions de Maré,
Nouvelle Caledonie: Les Eletok. Publications de
la Société des Océanistes 35. Paris: Musee de
Forestier, H. 1994. Contribution à la connaissance du
peuplement du Pacique Sud-Ouest. L’industrie
lithique des premiers mélanésiens de Nouvelle-
Calédonie: Etude du site de Naïa (Province Sud) et
quelques éléments de comparaison avec la région
Rocks, pottery and bird bones
de Koumac (Province Nord). Unpublished MA
thesis, Muséum National d’Histoire Naturelle.
—. 1996a. New Evidence of A Stone Tool Technology
in New Caledonia. Bulletin of the Indo-Pacic
Prehistory Association 14:68 -72.
—. 1996b. L’indust rie lithique des premiers mélanésiens
de Nouvelle-Calédonie. Journal de la Société des
Océanistes 102:71-89.
—. 1999. “Concept and Method in Lithic Production
during Lapita Period in New Caledonia: A
Technological Analysis,” in Le Pacique de 5000
à 2000 avant le présent: Suppléments à l’histoire
d’une Colonisation (e Pacic from 5000 to 2000
BP: Colonisation and Transformations). Edited
by J.-C. Galipaud and I. Lilley, pp. 345-361. Paris:
IRD Editons.
Guiart, J. 1963. Structure de la cheerie en Mélanésie
du Sud. Paris: Institut d’ethnologie.
Lagarde, L. 2004. Le matériel lithique du site
WKO013A. Premiers résultats et comparaisons
avec le matériel d’autres sites anciens de Nouvelle-
Calédonie. Unpublished MA thesis, Université
Paris 1 Panthéon-Sorbonne.
—. 2012. Peuplement, dynamiques internes et relations
externes da ns un ensemble géographique cohérent
de Mélanésie insulaire: lexemple de l’Île des
Pins en Nouvelle-Calédonie. Unpublished PhD
disser tation, Université Paris 1 Panthéon-Sorbon ne.
Lambert, P. 1900. Mœurs et superstitions des Néo-
Calédoniens. Nouméa: Nouvelle Imprimerie
Leenhardt, M. 1930. Notes d’ethnologie néo-
calédonienne. Travaux et Mémoires de l’Institut
d’Ethnologie. Paris: Institut d’ethnologie.
—. 1937. Gens de la Grande Terre. Paris: Gallimard.
Linnaeus, C. v. 1758. Systema Naturae, 10th edition
edition. Holmiae: Laurentii Salvii.
Maurizot, P. 2003. État des connaissances sur la
géologie de la Nouvelle-Calédonie. Géologues
Patouillet, J. 1873. Trois ans en Nouvelle-Calédonie.
Paris: Dentu.
Pisier, G. 1972. Kounié ou l’Île des P ins. Nouméa: Société
d’études historiques de la Nouvelle- Calédonie.
Sand, C. 1999. e Begi nning of Southern Melanesian
Prehistory : e St. Maurice-Vatcha Lapita Site, New
Caledonia Journal of Field Archaeology 26: 307-323.
—. 2001. “Changes in Non-Ceramic Artefacts
during the Prehistory of New Caledonia,” in e
Archaeology of Lapita Dispersal in Oceania: Papers
from the Fourth Lapita Conference, June 2000,
Canberra, Australia, Terra Australis 17. Edited
by G. R. Clark, A. J. Anderson, and T. Vunidilo,
pp. 75-92. Canberra: Pandanus Books. Research
School of Pacic and Asian Studies, Australian
National University.
—. 2 010. Lapita Calédonien: Archéologie d ’un premier
peuplement insulaire océanien. Col lection travau x
et documents océanistes 2. Paris: Société des
—. Editor. 2002. Walpole: Ha Colo, une île de
L’extrême Archéologies et Histoires. Les C ahiers de
l’Archéologie en Nouvelle-Calédonie 14. Nouméa:
Département Archéologie, Service des Musées et
du Patrimoine de Nouvelle-Calédonie.
Sand, C., a nd S. Bedford. Editors. 2010. Lapita: Ancêtres
Océaniens. Paris: Musee du quai Branly, Somogy
Editions D’Art.
Sand, C., J. Bolé, and A. Ouetcho. 2011. A Revision
of New Caledonia’s Ceramic Sequence. Journal of
Pacic Archaeology 2:56-68.
Sand, C., and A. Ouetcho. 1992. Bwé dé Ko -Trõ-Trè vã
Varè-Kè (Des rivières déviées par les ancêtres).
Premier inventaire archéologique de la commune
de Yaté, Province Sud de la Nouvelle-Calédonie. Les
Cahiers de l ’Archéologie en Nouvelle-Calédonie 1.
Nouméa: Département Archéologie, Service des
Musées et du Patrimoine de Nouvelle-Calédonie.
Sand, C., and P. J. Sheppard. 2000. Long Distance
Prehistoric Obsidian Imports in New Caledonia:
Characteristics and Meaning. Comptes-Rendus
de l’Académie des Sciences de Paris. Sciences de la
Terre et des Planètes 331:235 -243.
Louis Lagarde and André-John Ouetcho
Rocks, pottery and bird bones
... It appears to have been used in New Caledonia and other islands of southern Vanuatu during the same time period. Primary and secondary burials with no associated ornaments were only partially excavated at Tiouandé (New Caledonia), Hnenigec (Loyalty Island), Île des Pins (Lagarde and Ouetcho 2015;Valentin and Sand 2008) and Raowalai on Erromango (Valentin et al. 2011). ...
In this paper we present petrographic analyses of 68 Lapita pottery sherds excavated from the St. Maurice-Vatcha Lapita site located on the southeast coast of Île des Pins, New Caledonia and dated between 950 and 800 cal BC. Nearly two-thirds of these samples were without doubt made in the Diahot River valley region at the northwestern end of the main island (Grande Terre) some 400 km away from Île des Pins. Samples from another major geological region on the west coast of Grande Terre have also been identified, but no local production has been recognized so far. We propose that long-distance pottery transportation between the Diahot River valley settlers and those who inhabited Île des Pins may indicate: (1) an early establishment of Lapita settlements (so far undiscovered) in the Diahot region before occupation of Île des Pins; and (2) a persistent and preferred relationship between Île des Pins and the Diahot River valley. Other communities located on various parts of Grande Terre also established relationship with inhabitants of St. Maurice-Vatcha, but these do not appear to have been as intensive or long-lasting as those from the Diahot River valley. Surprisingly, Lapita people of the nearby Loyalty Islands – probably the earliest Lapita settlers of the entire New Caledonia archipelago – do not appear to have supplied pottery to those occupying Île des Pins during this time period.
Full-text available
The peopling of the western Pacific south of the Solomon Islands started with the spread of Austronesian populations more than 3000 years ago. The initial colonization of Remote Oceania is archaeologically linked to the Lapita Cultural Complex identified at more than 100 major sites from Island Papua-New Guinea to western Polynesia by a distinctive type of dentate-stamped decorated pottery and associated items. Until the last decade, Lapita sites of New Caledoniay the southernmost archipelago of Melanesia, had been poorly studied, leading to unsatisfactory conclusions about the characteristics and the length of the founding cultural complex. New excavations undertaken at the most southerly Lapita site of the Pacific, the St Maurice-Vatcha site on the Isle of Pines, shed new light on the matter. Although one of the first Lapita sites identified in the region and excavated by different teams over the years, its stratigraphy, chronology, and artifacts were not well understood. The new excavations have led to a precise understanding of the stratigraphy, and a large set of new 14C dates have firmly established Lapita chronology. The study of the archaeological material helps to test the proposal of a “Southern Lapita Province,” characterizing the earliest sites of southern Melanesia.
Full-text available
The Lapita Cultural Complex marks the first archaeological presence of human groups in the region of Remote Oceania, east of the main Solomon Islands. The spread of Austronesian seafaring communities from the Bismarck archipelago (Island Papua New Guinea) to the remote islands of Western Polynesia between 3300–2850 BP, has been characterised archaeologically by a period of shared cultural relations over long maritime distances. The significance of the discovery, in two of the earliest Lapita sites of New Caledonia, of obsidian flakes originating from the Talasea source in New Britain more than 2000 km to the north, is discussed in this paper.RésuméLe complexe culturel Lapita représente la première installation humaine en Océanie lointaine, à l’est des îles Salomon. L’expansion de communautés de navigateurs austronésiens de l’archipel de Bismarck, à l’est de la Papouasie Nouvelle-Guinée, jusqu’aux petites îles de Polynésie occidentale entre 3300 et 2850 BP, a été caractérisée comme une période de relations culturelles partagées sur de longues distances maritimes. La signification culturelle de la découverte — dans deux des plus anciens sites Lapita de Nouvelle-Calédonie — d’éclats d’obsidienne provenant géologiquement de la source de Talasea en Nouvelle-Bretagne, à plus de 2 000 km vers le nord, est discutée dans cet article.
La forêt de l'Ile des Pins
  • J.-F Cherrier
Cherrier, J.-F. 1986. La forêt de l'Ile des Pins. Bois et Forêts des Tropiques 211:35-46.
Mythes et Traditions de Maré
  • M.-J Dubois
Dubois, M.-J. 1975. Mythes et Traditions de Maré, Nouvelle Caledonie: Les Eletok. Publications de la Société des Océanistes 35. Paris: Musee de l'homme.
Structure de la chefferie en Mélanésie du Sud
  • J Guiart
Guiart, J. 1963. Structure de la chefferie en Mélanésie du Sud. Paris: Institut d'ethnologie.
Le matériel lithique du site WKO013A. Premiers résultats et comparaisons avec le matériel d'autres sites anciens de Nouvelle-Calédonie
  • L Lagarde
Lagarde, L. 2004. Le matériel lithique du site WKO013A. Premiers résultats et comparaisons avec le matériel d'autres sites anciens de Nouvelle-Calédonie. Unpublished MA thesis, Université Paris 1 Panthéon-Sorbonne.
Moeurs et superstitions des Néo- Calédoniens
  • P Lambert
Lambert, P. 1900. Moeurs et superstitions des Néo- Calédoniens. Nouméa: Nouvelle Imprimerie Nouméenne.
Notes d'ethnologie néocalédonienne. Travaux et Mémoires de l'Institut d'Ethnologie. Paris: Institut d'ethnologie. -. 1937. Gens de la Grande Terre
  • M Leenhardt
Leenhardt, M. 1930. Notes d'ethnologie néocalédonienne. Travaux et Mémoires de l'Institut d'Ethnologie. Paris: Institut d'ethnologie. -. 1937. Gens de la Grande Terre. Paris: Gallimard.
État des connaissances sur la géologie de la Nouvelle-Calédonie
  • P Maurizot
Maurizot, P. 2003. État des connaissances sur la géologie de la Nouvelle-Calédonie. Géologues 138:10-14.