Seasonal environmental practices and climate fluctuations in Melanesia. An assessment of
small island societies in Papua New Guinea and Vanuatu.
Frederick H. DAMON1
Carlos MONDRAGÓN2
This paper presents two exemplary case studies of indigenous adaptability to environmental change in
the region of the Western Pacific known as Island Melanesia. We draw on extensive experience of
local environmental knowledge and short- and long-term fluctuations in two small Pacific Islands’
societies, the island of Muyuw, on the northern side of the Kula ring (maritime Papua New Guinea),
and the Torres Islands, in the north of the Vanuatu archipelago. We offer a critical assessment of the
contemporary state of human-environmental relations in these communities, with special attention to
sea level rise and the El Niño Southern Oscillation (ENSO), in view of these being signature issues
within climate change discussions in Pacific Islands. We place these topics within the framework of
primary food production systems and local forms of guardianship and exploitation of forest and marine
resources. Throughout, our focus is on these communities’ reactions to observed climatic and
geological fluctuations over the past two decades: in the first case, in relation to the human
modification of forest resources in Muyuw, while the second case focuses on the shoreline dynamics
provoked by sudden and extreme sea level fluctuations as a consequence of seismic activity in
Vanuatu. We argue that the observed effects of these processes offer valuable data regarding the
specifics of human adaptation to climate change in small islands’ contexts. One of our key arguments
is that mid and long term environmental fluctuations have long been a part of Melanesian
engagements with the physical world, and have consequently given rise to coping strategies that are
inherent to traditional knowledge practices. We also seek to emphasize that the geomorphology of the
Melanesian islands –which tend to be small land masses that are often graced with multifarious soil
types and forest and marine resources– already possesses degrees of ecological adaptability which
appear to be absent from other idiosyncratic Pacific Islands’ contexts, most notably those of the lowlying atolls proper to Micronesia and Polynesia, which are often taken as exemplary of all small islands
around the world. It follows that assessments of climate adaptation across much of the Western
Pacific, and indeed large parts of Island Southeast Asia, must be approached and understood in ways
that are sensitive to the local differences between these environments and those of insular societies in
the Central and Eastern Pacific Ocean.
Keywords Climate fluctuations, sea level rise, indigenous adaptation, small island societies,
Melanesia
I. Introduction
Since 2008, with support from the US National Science Foundation, the authors have carried out
comparative research initiative undertaken with more than 15 colleagues who specialize in different
areas of the Pacific Islands, Asia and the Americas; our primary focus has been the flexibility of
traditional environmental knowledge in different parts of the Pacific Rim and Islands. This work has
1 Department of Anthropology, University of Virginia, USA
2 Centro de Estudios de Asia y África, El Colegio de México, MEXICO
1
been facilitated by closely collaborating with specific local actors. In the case of Vanuatu, this
involved the Vanuatu Cultural Centre and National Museum through its unique Extension
Fieldworkers’ Program, which is aimed at enabling the participation of local people in the
processes and output of scientific research (Bolton 1999, 2007). In maritime Papua New Guinea,
Damon has carried out over three decades of dedicated fieldwork, with an environmental focus
since 1991 (Damon 1998, 2005a, 2005b, 2008).
In this paper we present two case studies of indigenous adaptation to various forms of
environmental transformation in Island Melanesia. One of our main arguments is that approaches
to environments and societies in small islands contexts require careful consideration of local
familiarity with ecological change, with special attention to existing long-term trends in the adaptive
capacity of social and environmental milieux. Moreover, in both cases the anthropogenic nature of
key environmental spheres are critical to understanding the mutually constituted nature and
resilience of social and physical worlds. We approach adaptive capacity through a discussion of
factors that range from mid and long-term climate fluctuations to very severe, short term changes
in specific parts of the shoreline. The object of our discussion is twofold: to draw attention to the
heterogeneity of small island environments and provide some insight into the surprising range of
adaptation that local societies evince.
II. Changing Scenes in eastern Papua New Guinea
1. Background to Milne Bay Province, PNG
From the vantage point of nearly four decades of research focused on the northeast corner of
Papua New Guinea’s Milne Bay Province, two matters stand out concerning traditional practices
and knowledge, and experiences of climate change. One of these concerns experienced sea level
rise. For people who live by the sea –proximate to it and by means of it—these are significant
observations. The second involves observed alterations in complex understandings of weather
patterns. These entail perceived changes with respect to intra- and interannual oscillations
2
between sun and rain. Without any question the region of Damon’s concern was organized by the
sea and by rain/sun dynamics. Their experience of change, as is the case with the sea level, is no
small matter. Documenting these circumstances is the purpose of this section.
Muyuw, or Woodlark Island, is in the northeast corner of one of anthropology’s classic fields
of study, the Kula Ring made famous from the writings of Bronislaw Malinowski from the late teens
of the last century (e.g. Malinowski 1984[1922]). Matters of time were a principal focus of Damon’s
initial work (Damon 1982, 1990); they re-emerged as important concerns as his ethnobotanical
studies deepened from 1991 to the present. Those inquiries led him to learn about indigenous
peoples’ interest in the weather, and the experience of a shifting climate. He received worried
reports that older patterns increasingly rarely came to be for the first time in 1991, as did other
students of the region at the same time. The arrangements were understood to be oscillations of
sun and rain modeled on the behavior of stars, and the appearance of “big suns,” prolonged
periods of drought which we call El Niño Southern Oscillation. The first of these concerns intraannual patterns, the second interannual. People told him they were not working the way they
should; the early report has been sustained over the last twenty years.
Two external temporal matters might be raised here by way of an analytical introduction.
First, most of the inhabitants of Milne Bay Province derive from the Austronesian expansion into
the South Pacific beginning some 4000 years ago, an historical transformation which in this part of
Melanesia overlaid a human history at least 30,000 years in the making. Because of its geology
and climate variation this region is one of the more dynamic—“patchy” is the correct ecological
term-- ones in the world. A conclusion from this history is that for long the cultural forms of this
region have probably been closely aligned with changing environmental relations, in the seas, on
the lands, and through the atmosphere.
The second temporal matter concerns the present: Milne Bay Province has been on the
edge of Western Expansion since the first half of the 19th century. One recent transformation into
the modern is the transition from colonial-era miners with shovels and picks (from 1895-1940) to
3
massive petroleum-driven drills, earth removers and movers. Mass power rather than muscled
dexterity is the new cultural reality.3
2. Changing sea levels
Many local people speak about gradual but observable rises in the sea level. These are
observations not to be trusted completely because coasts are always and everywhere incredibly
dynamic locales (cf. Mei and Liu 1993; Klein et al. 1998; Haslett 2009: 131-154). But when Damon
told people in 2006 and 2009 that he did not recall seeing so many trees toppling into the water
they said they had observed the same thing. Young women who frequent shoreline coral reefs
looking for shellfish report that the water always seems higher and some reefs never become
exposed by the lowest of low tides.
But more convincing than these reports are those coming from small islands to the west,
Gawa, Kweywata and Iwa, and the south, Nasikwabw. Damon has not visited the first three since
the 1990s but reports from them say people are worried about their boat landings. Most of these
islands appear like pillars raised up out of the ocean 30-150 meters with flat, slightly concave tops
where people live. They are not in danger. Yet life is managed on these islands by the water
connections to elsewhere and higher water and hence more wave action is damaging their means
of connection. The situation on Nasikwabw, a sailing village to the south, is different because for a
hundred years or more the community lived not on the raised coral reef’s high plain but rather
nestled next to the sea along a semi-protected cove. By 2002, however, at least two families had
moved up on the top. And Damon was told in 2006 and 2009 that others had moved as well.
Waves regularly washed under houses, and it was increasingly difficult to protect boats dragged
onto large wooden rollers.4
3 KulaGold LTD now operates on the island; its committed and progressive thinking miners ridicule discussion of global warming.
4 1400CE is a significant pivot in the cultural history of this area and it is tempting to suggest that the reputed drop of the sealevel due to
the Little Ice Age would have enabled life on the beaches of Nasikwabw in a way perhaps not possible earlier (See Nunn, P.D. 2000,
4
The early colonial order sought to locate villages along shorelines, and many remain there
now. But before then various beach-hugging trees defined coast lines, probably a more intelligent
policy. If the recent experience of rising sea levels is sustained more villages may move to higher
locations, perhaps reverting to a wiser, traditional orientation.
3. A Time Chiasmus—Star and Tree times, and the new confusions…
The last two decades of my ethnobotanical research revealed an astonishing adaptation between
intricate cultural forms and climate/environment conditions. Space allows for these only to be
sketched here; for more complete treatment see Damon (1982, 1990,C1-2; 2005a). These facts
add new interest to the ‘problem’ of “primitive calendars,” for which this part of the ethnographic
world has been very significant. From Leach’s work(1950) on the Trobriand material (e.g.
Malinowski 1927, Austen 1939, [after Leach,1950]) Anthony Aveni, an astronomer who has been a
principle force in the ethno- and archaeo-astronomy movement, created a synthesis that is
suggestive for many of the world’s calendars, including the forms underlying the contemporary
Western construct (Aveni 2002[1989]).
The patterns here are very complex, not visible from a single research site nor evident over
the customary year or two of anthropological research.5
To begin, Muyuw has virtually the same set of lunar month names as found on the islands
of Iw, Kitava and the Trobriands, stretching to the west for some 160 kilometers. However, these
same names cover different slots of the solar year, and New Years are defined at different times
on different islands. Muyuw’s New Year starts in the eastern end of the island in roughly March,
clearly approximate to that equinox but not exactly fixed to it. In theory it is progressively declared
village by village moving west across the island; the organizing principle is the east to west
Nunn, P.D. and Britton, J.M.R. 200;1and Dickinson 2003). However, William Dickinson, personal communication, sees no proof for a 1
meter change in the Pacific sea level circa 1300-1400. These matters are very complex.
5 A well-intentioned ANU-based emergency trip through Milne Bay Province during the trough of the 1997-98 ENSO event arrived on
the small island of Yemga only to discover that its gardens were not sufficient to sustain the population. That was falsely attributed to
that ENSO event. Yemga is a sailing village never sustained by its gardens; its work was to maintain a sailing fleet that enabled regional
exchange, and it did that in exchange for garden work found elsewhere.
5
movement, not a fixed time interval. By the western end of the island, and the next island over,
Gawa, this paradigm dissipates, as if an orientation effectively dissolves. Iwa Island, exactly
between Muyuw and the Trobriands, thinks it starts the calendar that defines the Trobriand
sequence, and does so with the first full moon after the heliacal rising of the Pleiades. This is thus
after the 1st of June so proximate to the solstice then. The progression moves on to the next island,
and time districts in the Trobriands, by successive full moons so that while there is a spatial
component to the sequence, the data argues that the form is fixed by the temporal intervals the
moon creates. Vakuta, at the Trobriand’s southern extremity and the last place that operates this
calendrical system, declares its new year in September or early October. The whole sequence thus
begins roughly with one equinox, goes to the intervening solstice, and ends with the next equinox.
This is thus an intra-annual ‘time system,’ mediated by a cultural transformation, which is
part of a regional system organizing a set of sequential activities that transcend a solar year (See
Damon 2005b for discussion of the ‘chaos’ systemic properties of these relationships).
This fact is part of the genius of this highly developed regional network. For now it is
enough to observe that the positioning of the lunar month names about the solar year makes the
eastern and western sides of this sequence almost mirror images of one another. It is important to
note then that most of Muyuw and most of the Trobriands plant their main annual crops, yams,
about the same time, in December, and harvest them after April and closer to June and July. The
little islands in between invert this almost exactly, harvesting in the Austral summer, planting in the
winter.
Times and activities vary here because people expect intra-annual patterns to be
unpredictably upset by interannual patterns, the droughts of ENSO events. The little islands are
little, which means they never absorb much rainwater, and they are high, which means the ground
surface is way above the underground water lens. With a drought of any length they lose all their
crops. And have to have them replaced by the peoples to their east. It turns out that the people on
these small islands tend not to know very much about their root crops. While this first puzzled me,
6
it made sense after I realized they regularly lost those crops, depended on others (with deeper
knowledge) to replace what they lost, and concentrate instead on quite extraordinary sets of trees
that, because of their deeper roots, more easily withstand all but the worst of ENSO-generated
droughts. Iwa Island people maintain hundreds of nut and fruit trees while those on Gawa and
Kweywata maintain trees used to construct the highest class of outrigger canoe used in the
interisland communication system.
Trees mediate between the annual crops of yams and taro, and the interannual relations to
a major degree effected by ENSO circumstances. Does this combination synthesize the original
inhabitants of Melanesia and eastern Indonesia and the Austronesian expanse that brought the
East Asian root crops to the region?
But let us return to the formal patterns. The working model runs like this. A set of
approximately twelve starts/constellations are watched for their heliacal risings and settings.
Metaphorically they are conceived to be like “Important Men”—which means men well established
in interisland networks. And like these men, when they die, when they set in the west in the
evening sky, they release a power which results in wind and rain (Some people correlated the
death of one of Damon’s great teachers in late 1996, with a following cyclone and the ensuing
ENSO of 97-98.). That period is then followed by a week or more of sun. The oscillation enables
the swidden system, cutting a forest, burning it, then planting it. Sun is needed to power growth but
also to dry a cut forest enabling burning, planting and harvesting. But that weather set has to be
broken by the equally necessary rain. The annual movement of the stars models the oscillation
between sun and rain.
Added to this patterning is a variance in ideal fallow periods. These realize the possibilities
between no gardens just forests and gardens and forests grown together. The structure is too
complex to completely describe here. Suffice it to say that the small islands between Muyuw and
the Trobriands mix gardening and fallowing by planting trees in with their crops. One place on
Muyuw is famous for the other limit, no gardens at all, so of course no fallowing, just forests. In
7
between these extremes, eastern Muyuw ideally uses a short fallow cycle—so 10-15 year intervals;
central Muyuw uses middle length fallows—20-40 years; and western Muyuw ideally employs a
long fallow.
These differences are regionally, not just locally, significant. Muyuw believe that older
forests keep the ground cooler and wetter. The convenient fact for these practices being located in
western Muyuw is that it helps make their yam gardens virtually impervious to ENSO extremes,
and this right next to the small islands for whom even the shortest droughts are a disastrous for
root crops. When I returned to Muyuw in 1998 after the extremely severe ENSO event was over all
across Muyuw people were sending yams, as food and seed, to the western islands. However,
only in western Muyuw did people tell me that the ENSO had little effect on them, and only there
did they say they were sending so many yams west that they couldn’t count them.
This diverse set of fallowing generates the trees appropriate for producing the inter-island
outrigger canoes. As noted the canoes are made in whole in Gaw and Kweywata. However, those
islands lack the best trees for certain parts in general and some of the best replacement parts in
particular. These come from Muyuw—so masts from the part of the island that does not garden;
the best outrigger floats from central Muyuw by means of its ideal middle-length fallow; large
numbers of really small trees useful for making outrigger platforms from the early fallowing
practices in the east. The land and its products have been sculpted to generate the materials found
necessary for the region’s unique outrigger boat form. This is not coincidental because the chaotic
conditions of sailing experiences, i.e. these boats, are models for the chaotic weather patterns
these people customarily experience. A boat is supposed to be made in a certain way but each
boat has unique properties that have to be individually attended to, as do the relations between the
islands. It is no accident that the idea of a boat is imposed over most forms—villages, gardens,
whole islands—and social relations (See Damon 1998, 2008).
Superimposed over the annual movements modeled by the stars are interannual
oscillations produced by what we call ENSO. These create oscillations between longer periods of
8
drought and longer periods of rain. This pattern complements the annual cycle because with
respect to it sago orchards are managed and individual sago trees harvested. Sago trees are palm
trees domesticated somewhere between eastern Indonesia and eastern New Guinea; they are this
part of the world’s contribution to the human carbohydrate stock (Ellen 2004, 2006). From a mature
tree carbohydrates are obtained by pulverizing the trees’ pith and by water and hand separating
the sugars from the cellulose. A flour-like material results. Although some parts of this region use
sago as their mainstay, in this part of PNG the material forms a condiment to every ritual,
exchange item for the interisland set of regional dependencies, supplement for the low spot in the
yam and taro cycle—in the Austral summer—and emergency food for severe ENSO droughts. As
part of the practice’s management system orchards are juxtaposed to meadows found in the high
forest. These meadows are created by fire during ENSO droughts—otherwise trees would infill
them. They produce important materials for tying bodies and boats. And I suspect run-off from the
meadows enhances nutrient availability in the orchards, some of which are contiguous to the
meadows, but all of which are downstream from them. Beyond that hypothesis, I know people coordinate sago orchard/meadow relations because “wild-pigs,” avidly hunted and eaten, take refuge
in the meadows and eat left-over sago purposely left for them.
The garden/fallow and sago orchard/meadow relation is a coordinated system partly
fashioned together by means of the way social relations are represented in each form. Gardens
contain models of the culture’s gender and clan system, a set of continuous complementary
replacement; sago orchards model subclan relations since the pattern of growth of individual trees
is equated with the way the culture’s matrilineal units are reproduced by a daughter taking the
place of her mother.
The relationships between the higher order ENSO oscillations and the lower order star
patterns created the dynamic around which this cultural system organized itself over the last 2000.
But if it describes an oscillation of times around which the culture created itself, it is not what
people experience, or talk about, now. Increasingly frequent prolonged periods of rain all but
9
eclipse the much shorter oscillation between sun and rain configured by the twelve or so stars. So
increasingly frequently both burning and harvesting extremely become difficult, violating what is
understood as a necessary order to gardening practices.6 And it may be that ENSO dynamics
enhanced by the effects of global warming will make the encompassing pattern truly dominant.
What cultural shape that will bring is for the future to determine. And that future will also be
determined in part by planned mining activities which are likely to be conceived independently of
the natural patterns with respect to which this set of regional determinations organized itself over
the last two millennia.7
II. Torres Islands8
The Torres group is made up of six small islands located at the northern border of the Vanuatu
archipelago; this microarchipelago currently sustains a population of just over 1,000 people
scattered over 8 coastal settlements, and comprises two distinct Austronesian languages.9 The
following paragraphs summarize the climatological and environmental conditions of this milieu and
focus on the sudden changes wrought on a specific area of the local coastline as a result of regular
seismic activity. We argue that the overall changes to the local shoreline – especially in relation to
soil quality, vegetation growth and hydrodynamics –, as provoked by extreme seismic uplift and
downlift, offer a unique and informative example of the long-term adaptability that is present in both
the human population and the observed coastal milieu of these islands. We contend that this
inherent adaptability can also be made extensive to other parts of the Western Pacific.
6
Outbreaks of malaria are now common. Vast quantities of anti-malaria medicines help create a population island resources cannot
sustain.
7
In the mid-1990s the people exploring mining possibilities on the island figured they would have to construct their new landscapes
based on hypothetical 20-year flood plans. Damon told them in 1996 that there had been 2 five-hundred year floods on the Mississippi in
the last 4 years.
8 For the past 12 years Mondragón has been working closely with various local actors in an effort to gather and produce a systematized
data base of local environmental knowledge. Importantly, these efforts have been supported and mediated by the Vanuatu Cultural
Centre through their Extension Fieldworkers Programme, which has been widely recognized to be one of the most successful schemes
for bringing together the methods and aims of anthropological work with the concerns and interests of local communities in the South
West Pacific.
9
Known as Lo-Toga (herafter L-T) and Hiw, these local languages belong to the East Vanuatu languages, a subgroup of the Oceanic
family The maritime frontier of North Vanuatu and the South West Solomons represent an extended cultural borderland where the
influences of Maritime Melanesia and Eastern Polynesia meet and overlap. Consequently, there are a number of shared cultural
features that link some of the Austronesian-speaking Melanesian and Polynesian societies of this region.
10
1. The physical environment
The landmasses that make up the Torres group are the product of ancient coral uplifts which range
in average elevation between 150 and 360m.a.s.l. (Galipaud 1996). The Torres Islands are
covered by dense vegetation whose flora and fauna are proper to the northern Vanuatu rainforest
ecoregion, which extends into the neighbouring South West Solomons (CSIRO n.d.; Wheatley
1992: 6-8). As in the vast majority of ni-Vanuatu communities throughout the archipelago,
horticulture and agroforestry are the main activities towards which people’s daily subsistence
efforts (dominated by yam gardens) and cash cropping (primarily copra plantations) are directed
(Weightman 1989: xix-xx). It is worth noting that horticulture in Vanuatu is characterized by
multicrop gardens whose extreme productivity gives rise to bountiful yam harvests –often referred
to as “subsistence plus”– which are vital to the traditional exchange systems on which much of
local cultural forms are predicated (Weightman 1989: 29; Rio 2007: 105-131). However, the
islands’ low elevations mean that large parts of the Torres –namely their coastal regions, where the
principal food-producing gardens and the totality of present-day villages are located– are at
potentially greater threat from sea level rise than other Vanuatu islands (e.g. Reti 2007). The small
surface area heightens this initial appearance of vulnerability. However, as we show, these
apparent levels of vulnerability are mitigated by a series of factors, including the heightened
flexibility of settlement and gardening patterns, as well as the variability of fertile soils.
2.Soils and garden distribution
Torres islanders cultivate at least two major types of garden, one hilltop and one coastal, per
annum. The primary crop of Torres Islands’ horticulture is the yam. Although Dioscorea alata is the
predominant type there are a great many yam varieties, each with its own name and associated
quality, which share any given plot of land with other garden plantings. Equally, there is a
considerable body of botanical knowledge which mostly concentrates on leaves and vines. For
11
reasons of space we will not go into this ethnobotanical corpus here. Most nucleated families (L-T:
metaviv) plant and tend anywhere from 3 to 7 different gardens per year on different types of soil.
The location of a garden and its associated soil type is the primary marker by which any
given plot is defined in the local language; in other words, the names of garden types are
dependent primarily on the kind of soil and height above sea level at which they are situated.
Torres islanders recognize two major soil types, ‘hard’ and ‘soft’, and at least four relevant
subtypes, in relation to horticultural activity and specific garden plantings, (Fig. 1; see Weightman
1989: 9-10, for a general listing of soil types across Vanuatu). Thus, the categories ‘hilltop’ and
‘coastal’ are themselves generalizations which encompass at least 6 different garden/soil
subtypes. Lowland garden soils are said to be more productive and are planted and harvested first;
they are the main providers of daily food year-round. Hilltop gardens are, however, the mainstay
for surplus produce that is directed at ceremonial exchange and long term storage, and are
therefore critical to the cultural expression and social reproduction of these communities.
Name
Lete li venie
(“earth garden”)
Lete Lo
(“soft garden”)
Quality
Subtype
Kind
Tenë wetagë
Most
common
Black topsoil, red
underneath
Ten lave
Large
garden
Grey soil, volcanic
(v. fertile)
Ten mevë
‘Heavy’
garden
“hard”/hilltop
soils
“soft”/coastal
soils
Soil type
They are planted first and constitute the more
accessible food producing gardens.
Fig. 1. Torres Islands’s soil types. Carlos Mondragón 2011
Average fallowing times in traditional horticultural practice average between 7 and 20 years
per plot, although at present most fallowing times have been reduced to between 3 and 5 years.
12
This reduction follows more than three decades of successive advice emanating from external
agricultural experts associated with the central government and various international organizations
and development outfits, which have tended to motivate greater productivity by reducing fallowing
times and intensifying the planting and harvesting cycles of garden plots.
Importantly, plots are distributed irregularly across the surface area of the islands, following
a form of territorial rights which allocates discontinuous plots of fertile soil to any given family. The
irregular, discontinous distribution of plots, in turn, follows from patterns of matrilineal inheritance in
which patches of garden land are constantly shifting in relation to changing marriage alliances. The
result of this is that each extended family unit (metaviv) belonging to larger matrilineages (L-T:
tutumwa) owns a patchwork of shifting plots that are located in different parts of every island.
Because it implies a distribution of environmental risk, this scattered pattern of food
producing sites is critical to enhancing human and ecological capacities to cope with climatic
fluctuations such as drought, excessive rainfall or future sea level rise.10 In this respect, it can be
clearly stated that local forms of social organization, i.e. the Torres kinship system and its
associated forms of inheritance, settlement patterns and gardening times, in conjunction with
knowledge of varieties and availability of major fertile soil types, constitute a key reference for
outsiders wishing to understand Torres islanders’ capacity to adapt to climate change in the near
future.
3. Reefs, tides and hydrodynamics
The ocean around the Torres Islands is mostly dominated by deep-sea environs, given that
the land drops sharply into the depths. This means that its reefs are squeezed into a very narrow
vertical ecological niche that lies between the shoreline and the colder seawater below. In this
10
This diversified pattern of horticulture also has a profound effect on the nature and distribution of nonhorticultural and non-food producing vegetative growth. Local people are keenly aware of this, because
patterns of distribution of specific tree and plant species also impinge on the renewal of soil fertility.
13
regard, the reef systems of the Torres, which provide the bulk of quotidian protein consumption,
are one particular environmental sphere that is highly vulnerable to extreme changes in sea levels
and in average sea surface temperatures.
In addition to reefs there are at least two major coastal sites dominated by mangrove
shallows which concentrate much of the high-productivity coastal and maritime tenure of the Torres
islanders.11 Torres islanders’ quotidian marine tenure, and the regular displacement of land and
sea animals within these mangrove ecosystems, are intimately tied to tidal levels, which in turn
follow the changing position of the moon in the sky. This may help to explain the existence of a
highly developed set of indigenous categories for tides (Fig. 2).
Category
Liave
High tide
Met
Low tide
Name
Type
Liave li vu liave
Liave me tave
Liav revrev
Liave li lowate
Liavi ihar
Liave meren
Nu metmet
Met metave
Met mëren
Met revrev
Met melige
Met mëgagë
Met mëgagë metavu toten
Beginning of any high tide
Morning tide
Evening tide
Midday tide
Highest tide of the year
High tide during daylight
Beginning of any low tide
Morning tide
Midday tide
Evening tide
New moon
Full moon
Low tide during the day and
during waning quarter moon
Fig. 2 Torres Islands’ tides. Carlos Mondragón 2011
The highly organised body of ecological knowledge presented here demonstrates that
Torres people pay acute attention to those aspects of the environment that most closely influence
their lives – for, in addition to determining the various times that are dedicated to harvesting the
sea, the tides also affect the freshwater lens under the islands and thereby the soil moisture
11
By contrast, deep water fishing is generally reserved for periods of sustained maritime and climatological calm,
during the Austral summer.
14
content of their lowland (coastal) yam gardens, which, as indicated above, are the mainstay of
subsistence horticulture.
4. Regular (annual) climatic fluctuations
The climatology of the Torres follows an equatorial monsoonal regime that is driven by two
prevailing wind seasons common to the South West Pacific. This regime is dominated by the South
Easterlies (from May to November) and the irregular North and North Westerlies, which are typical
of the cyclone season (from December to April). Wind, however, is not simply wind to the Torres
islanders. Each wind, known generically as lang, possesses a certain mena (Oc: mana; generative
power over living things) and is associated with a cardinal direction. The areas from where each
lang arrives constitute part of a complex (absolute, horizon based) system of orientation by which
Torres islanders express location and movement within the overall seascape. In addition, each
wind-related area is associated with specific meteorological and spiritual forces that lie beyond the
visible horizon and are subject to manipulation by ritual specialists (viz. weather ‘magic’).
Wet and dry seasons are determined by prevailing wind patterns, and, by extension, so is
the annual planting cycle of the primary food gardens, which are dominated by the cultivation of the
yam.12 The Torres people register all of the above fluctuations by a set of temporal markers which
Mondragón has elsewhere reconstructed in the form of a lunar ‘calendar’ (Mondragón 2004).
This temporal framework, or time system, is not simply a mnemonic device in the abstract,
but is rather a schematic representation of the ways in which Torres people conceptualize their
broader lived world, insofar as they do not conceive themselves to be singular (unique) entities
within it (Mondragón 2009a); hence, the spirits of the dead are understood to inhabit the same
spaces as those of the living, as are various other kinds of forces and non-human spirits. Each
component of the environment, both the seen and the unseen, requires specific ways of relating to
12
“...the islands of Vanuatu are either predominantly yam growing or taro growing. The wet northern and eastern
islands are mainly taro growing, particularly in their upland areas, with yam cultivation largely restricted to coastal
regions.” (Weightman 1989: 29)
15
and evoking creative (or destructive) forces that have a direct influence on human, non-human and
ancestral presences. These ways of engaging with the environment encompass both practical dayto-day knowledge (horticulture, marine tenure) and specialist knowledge (weather magic,
garden/fertility magic, etc.).
It is important to recognize that these complex bodies of integrated knowledge are not
systematized. They are not taught in formal settings, but transmitted in bits and pieces by different
specialists in different contexts. They are therefore living, shifting bodies of knowledge which
remain extant insofar as they relate to the most relevant of local activities, namely food production
and ritualised exchange.
5. ENSO, seismicity and changing sea levels
In the past few years the authors have been exploring mid-term and long term environmental
flucutations in an effort to break free from the constraints presented by thinking of local Melanesian
environments in terms of annual (solar) cycles and seasonal changes. The results of this
exploration in the context of the Torres Islands reveal that there may be a certain synchronization
of ENSO-related (7 to 8 year) periods of drought and above-average rainfall which lead to
increased food production that is tied into important ritual activity, namely, ceremonial feasting
related to special status-alteration rituals that only take place once or twice during the average
lifetime of a person (Mondragón 2009b). Specifically, ENSO-related periods of drought tend to be
related with little to no production of ceremonial yams and kava, while wetter years produce the
contrary effect.13
In addition to regular climatic fluctuations, the Torres Islands are subject to different types of
longer term and even irregular fluctuations which have a direct impact on the local environment.
13
The insight for this information was made possible by closely following the agricultural and ceremonial cycle of the
Torres islands for over a decade. Importantly, these observations were greatly enhanced by local people’s active
engagement in Mondragón’s work on longer term cycles. The 1997-98 Niño provided a baseline for comparing nonENSO related years to ENSO-related weather conditions up to the present day.
16
Chief among these are hurricanes and earthquakes. Quakes regularly provoke localized but often
dramatic transformations to the shoreline of the islands.14
The most recent large earthquakes occurred between 1997 and 2008. In 1997 an 8.0
magnitude quake uplifted Linua island by almost 4 metres, while at the same time it pulled down a
part of the coast on the opposite side of the mangrove lagoon separating this piece of land from
Lo. The result of this dramatic shift was that several hundred metres of coconut plantations on Lo
island gradually became waterlogged and died off. Starting in 2001, various external visitors
representing different environmental agencies began to speak about climate change and sea level
rise to local people. In the absence of a direct correlation between the ’97 quake and the apparent
sinking of their island, people on Lo began to associate the death of their shoreline plantations to
the effects of global warming. This in turn gave rise to a follow up visit by the South Pacific
Regional Environmental Programme (SPREP) in 2003, from which arose the (mistaken) notion that
the Torres –and especially the island of Tegua, whose shoreline was also affected by the
earthquakes and accompanying tidal surge– were experiencing accelerated sea level rise (see
Nakalevu et al. 2005: vi, and SPC 2009).
In 2008, however, a 7.9 magnitude quake shifted Lo and Linua islands again, this time
lifting the coast of Lo by at least three metres. Consequently, the waterlogged ground on Lo has
dried up and new vegetation – mostly mangrove shoots – has begun to return to this area.
The broader lesson here, as transmitted by various local people, is that they are used to the
shoreline and even parts of the lowlands on the Torres experiencing sometimes dramatic shifts in
appearance. These shifts often have extreme consequences for lowland soil fertility, vegetative
growth and horticulture, in addition to sometimes provoking violent changes to local hydrodynamics
(within the mangrove lagoon, for instance). But on the whole, over their 3,000 year history local
14
The uplift that created the Torres group is an immediate product of the deep-sea Torres Trench. Located a mere 4
kilometres offshore, at 6,000 m.b.s.l., this area marks the seismically active boundary between the Australasian and
Pacific Plates. Earthquakes and tsunamis have been a continuing fact of life for local people for millenia.
17
islanders have learned to adapt to such abrupt transformations in coastal dynamics. This
adaptability is further enhanced by the constantly shifting nature of local settlements.
To summarize, it appears that patterns of mid-term drought and excessive rainfall are more
serious than sea level rise in this part of Island Melanesia. Given the general dryness of the climate
and the soils of the Torres, local gardens tend to be less productive than those located on larger,
volcanic islands which are subjected to far higher levels of rainfall. In this regard the most
important impact on local communities from global warming over the coming years is likely to come
in the form of diminishing garden returns, rather than changes to the local shoreline. It is here, we
believe, that the focus of extra-local schemes for mitigation and adaptation must concentrate.
Conclusion
The two case studies presented here offer a diversity of knowledge practices, adaptive capacities
and environmental resilience in relation to the changing ecological milieu of the Western Pacific
from which it is possible to draw a number of important conclusions. First, as is the case for other
eco-regions, it is clear that any comprehensive approach to the issue of adaptation to climate
change in small island environments has to be particularly sensitive to local conditions – both
natural and social (cf. Cronin et al. 2004; Mercer et al. 2007). Only then can relevant insights and
strategies be drawn in order to best address future forms of adaptation at a grassroots level
(Pernetta 1992; Moran 2006: 5). In this regard, there are clearly important differences not only
between the Melanesian environments discussed herein and those of other small island contexts –
most notably Polynesia and Micronesia– but also between the two societies and island groups at
issue.
On one hand, the island of Muyuw and the Kula Ring of Papua New Guinea constitute an
extremely dynamic climatological environment, one in which patterns of drought and rainfall have
had a profound impact on the human stewardship (and modification) of the islands’ vegetation and
soils. For their part, while evincing some of the processes related to anthropogenic landscapes, the
18
Torres Islands are situated in a very highly active seismic region which provokes constant, violent
shifts in shoreline ecologies and hydrodynamics. In both cases, the climatological and geophysical
factors at work are different, if comparable. Nevertheless, the shared human factor is, in both
cases, key to understanding the combined and enhanced adaptive capacities that both people and
territory have generated over several millenia.
The forms of adaptation that we have reviewed here speak to a slightly more positive
outlook for the possibilities of adaptation to sea level rise in various small island locales. In this
regard, we would draw attention to the fact that even such an apparently destructive and seemingly
unstoppable process as the disappearance of current shoreline environments in many small island
regions of the world need not necessarily fall prey to catastrophist projections (Barnett 2001). Our
case studies offer a powerful example of the resilience and adaptability of humans and
environments even in cases that seem extreme; indeed, we are tempted to proffer the additional
argument that these Melanesian communities appear to be far more likely to successfully adapt to
abrupt climate change than the encompassing, globalized societies and institutions that are
seeking to help them in this process of environmental transition.
REFERENCES
Austen, L. (1939). The Seasonal Gardening Calendar of Kiriwina, Trobriand Islands. Oceania 9,
237-253.
(1950) A Note on Dr. Leach’s ‘Primitive Calendars.’ Oceania 20, 333-335.
Aveni, A. F. (2002). Empires of time : calendars, clocks, and cultures. Boulder: University Press of
Colorado (revised from 1989 edition).
19
Cronin, S. J., Gaylord, D.R., Charley, D., Alloway, B.V., Wallez, S., Esau, J.W. (2004) Participatory
methods of incorporating scientific with traditional knowledge for volcanic hazard management on
Ambrym Island, Vanuatu. Bulletin of Vulcanology 66(7): 652-668.
CSIRO (undated), The Environments of Vanuatu. A classification of the natural resources of
Vanuatu and their current use as determined from VANRIS. Brisbane: CSIRO, Queensland Dept.
of Primary Industries Forest Service, Australian International Development Assistance Bureau.
Barnett, J. (2001). Adapting to Climate Change in Pacific Island Countries: The Problem of
Uncertainty. World Development 29(6): 977-993.
Bolton, L. Introduction (1999) In Fieldwork and Fieldworkers in Vanuatu (Oceania, Special Issue)
70:1-8
---------- (2006) The museum as cultural agent: the Vanuatu Cultural Centre extension worker
program. In South Pacific Museums: Experiments in Culture, C. Healy and A. Witcomb (ed.).
Melbourne: Monash University ePress.
Damon, F. H. (1982). Calendars and Calendrical Rites on the Northern Side of the Kula Ring.
Oceania 52(3), 221-239.
----------(1990) From Muyuw to the Trobriands: Transformations Along the Northern Side of the
Kula Ring. Tucson: University of Arizona Press.
----------(1998) Selective Anthropomorphization: Trees in the Northeast Kula Ring. Social Analysis,
42(3):67-99.
---------- (2005a) The Woodlark Island Calendar: Contexts for Interpretation.” In Songs from the
Indigenous Sky: Indigenous Astronomical and Cosmological Traditions of the World, Von Del
Chamberlain, J.B. Carlson and M.J. Young (ed.), Bognor Regis: Ocarina Books, pp. 348-357.
20
----------(2005b) ‘Pity’ and ‘Ecstasy’ The Problem of Order and Differentiated Difference Across
Kula Societies. In On The Order of ‘Chaos’. Social Anthropology & the Science of ‘Chaos’ M.
Mosko and F. Damon (ed.), Oxford: Berghahn Books, pp.79-107.
----------(2008) On the Ideas of a Boat. From Forest Patches to Cybernetic Structures in the
Outrigger Sailing Craft of the Eastern Kula Ring, Papua New Guinea.” In Beyond the Horizon.
Essays on Myth, History, Travel and Society. C. Sather & T. Kaartinen (ed.) Studia Fennica
Anthropologica 2. Helsinki: Finnish Literature Society. pp. 123-144.
Dickinson, W.R. (2003) Impact of Mid-Holocene Hydro-Isostatic Highstand in Regional
Sea Level on Habitability of Islands in Pacific Oceania. Journal of Coastal Research 19(3):
489-502.
----------(2004) The distribution of Metroxylon sagu and the historical diffusion of a complex
traditional technology. In: Smallholders and Stockbreeders: History of foodcrop and livestock
farming in Southeast Asia, P. Boomgaard and D. Henley (eds.). Leiden: KITLV Press, pp. 69-105
---------- (2006) Local Knowledge and Management of Sago Palm (Metroxylon sagu Rottbeoll)
diversity in south Central Seram, Maluku, Eastern Indonesia. Journal of Ethnobiology 26(2): 258298.
Galipaud, J.-F. (1996). Recherches archéologiques aux îles Torres. Premiers résultats de la
mission réalisée du 9 au 23 Novembre 1996 dans les îles de Metome, Tegua et Toga. Port Vila:
ORSTOM.
Haslett, S.K. (2009). Coastal Systems. 2nd edn. London: Routledge.
Klein, R.J.T., Smit, M.J., Goosen, H. and Hulsbergen, C.H. (1998). Resilience and Vulnerability:
Coastal Dynamics or Dutch Dikes?. The Geographical Journal, 164(3), 259-268.
21
Leach, E.R. (1950). Primitive Calendars. Oceania 20(4): 245-262
Malinowski, B. (1984 [1922]) Argonauts of the Western Pacific Prospect Heights, Il.:Waveland
Press.
----------(1927) Lunar and Seasonal Calendar in the Trobriands. Journal of the Royal
anthropological Institute 72: 203-215.
Mercer, J., Dominey-Howes, D., Kelman, I., Lloyd, K. (2007). The potential for combining
indigenous and western knowledge in reducing vulnerability to environmental hazards in small
island developing states. Environmental Hazards 7(4): 245-256.
Mondragón, C. (2004). Of Winds, Worms and Mana: the Traditional Calendar of the Torres Islands,
Vanuatu. Oceania 74(4): 289-308.
---------- (2009). A Weft of Nexus: Changing Notions of Space and Geographical Identity in
Vanuatu, Oceania. In Boundless Worlds. An Anthropological Approach to Movement, ed. P.W.
Kirby. Oxford: Berghahn, 2009, pp.115-133.
Moran, E. (2006). People and Nature. An Introduction to Human Ecological Relations. London:
Blackwell.
Nunn, P.D. (2000) Environmental catastrophe in the Pacific Islands about AD 1300.
Geoarchaeology, 15: 715-740.
Nunn, P.D. and Britton, J.M.R. (2001) Human-environment relationships in the Pacific Islands
around AD 1300. Environment and History 7: 3-22.
22
Mei, C.C. and Liu, P.L.-F. (1993). Surface Waves and Coastal Dynamics. Annual Review of Fluid
Mechanics, 25, 215-40.
Nakalevu, T., Carruthers, P., Phillips, B., Saena, V., Neitoga, I., Bishop, B. (2005). Community
Level Adaptation to Climate Change: Action in the Pacific. Proceedings of the Regional Workshop
on Community-Level Adaptation to Climate Change, Suva, Fiji: 21-23 March 2005. Suva, Fiji:
Secretariat of the Pacific Regional Environment Programme (SPREP) and Department of
Environment Fiji.
Pernetta, J.C. (1992) Impacts of climate change and sea-level rise on small island states: National
and international responses. Global Environmental Change 2(1): 19-31
Reti, M.J. (2007). An Assessment of the Impact of Climate Change on Agriculture and Food
Security in the Pacific. Apia, Samoa: Food and Agriculture Organization of the United Nations,
Subregional Office for the Pacific Islands.
South Pacific Community (SPC) (2009). Policy Brief 7/2009
Weightman, B. (1989). Agriculture in Vanuatu. A Historical Review. Cheam, Surrey: The British
Friends of Vanuatu.
Wheatley, J.I. (1992). A Guide to the Common Trees of Vanuatu, With Lists of their Traditional
Uses and ni-Vanuatu Names. Port Vila: Vanuatu Department of Forestry.
23
24