Times of Plenty, Times of Less: Last-Millennium Societal Disruption in the Pacific Basin

Times of Plenty, Times of Less: Last-Millennium SocietalDisruption in the Pacific Basin

Patrick D. Nunn & Rosalind Hunter-Anderson &

Mike T. Carson & Frank Thomas & Sean Ulm &

Michael J. Rowland

# Springer Science + Business Media, LLC 2007

Abstract During the last millennium in the Pacific Basin(islands and continental rim) there was a marked contrastbetween ‘times of plenty’ and ‘times of less’ for its humansocieties. This contrast is attributable to climate and sea-level variations, notably the Medieval Warm Period (A.D.700–1250) and the Little Ice Age (A.D. 1350–1800)separated by a time of rapid cooling and sea-level fallcalled the ‘A.D. 1300 Event.’ Outlines of the times of plentyduring the Medieval Warm Period and the times of lessduring the Little Ice Age are given, supported by a numberof examples. These confirm a general picture of societalcollapse as a result of the A.D. 1300 Event. Well-dated

human responses to the A.D. 1300 Event (establishment offortified settlements, end of ocean voyaging) allow links topotential nonhuman causes to be strengthened. Althoughmore data referring to both (natural) changes and theirhuman effects are needed, a conclusion involving environ-mental determinism is inescapable.

Key words Pacific Basin . climate change . sea-levelchange . societal collapse . environmental determinism

Introduction

Within the last millennium there was a marked contrast inmany Pacific Basin societies between what might be termed‘times of plenty’ during the Medieval Warm Period(approximately A.D. 700–1250) and ‘times of less’ duringthe Little Ice Age (approximately A.D. 1350–1800). Thetransition between the Medieval Warm Period and the LittleIce Age has been named the ‘A.D. 1300 Event’ and has beenidentified as a time of rapid cooling, sea-level fall, andcultural change. Specifically, it is clear that the societaldisruption associated with the A.D. 1300 Event was duelargely to a massive and rapid reduction of the foodresource base on which many Pacific communities hadcome to depend during the preceding Medieval WarmPeriod. In turn, this food reduction can be plausibly linkedto a number of environmental changes arising from rapidcooling and sea-level fall (Nunn, 2000, 2003; Nunn andBritton, 2001; Kumar et al., 2006; Masse et al., 2006).

This paper is not concerned with the cause(s)—near ordistant—of societal changes resulting from the A.D. 1300Event so much as the changes themselves. These changesare of interest because, by understanding their nature andtiming, more can be learned about the processes connecting

Hum EcolDOI 10.1007/s10745-006-9090-5

P. D. Nunn (*)School of Geography, Faculty of Islands and Oceans,The University of the South Pacific, Suva, Fijie-mail: [email protected]

R. Hunter-AndersonMicronesian Archaeological Research Services (MARS),P.O. Box 22303 GMF, Agana, GU 96921, USA

M. T. CarsonInternational Archaeological Research Institute, Inc. (IARII),2081 Young Street, Honolulu, HI 96826, USA

F. ThomasPacific Studies Program, Pacific Institute of Advanced Studies,The University of the South Pacific, Suva, Fiji

S. UlmAboriginal and Torres Strait Islander Studies Unit,The University of Queensland,Brisbane, Queensland 4072, Australia

M. J. RowlandCultural Heritage Coordination Unit,Department of Natural Resources and Mines,Locked Bag 40, Coorparoo DC,Queensland 4151, Australia

the two. This is an area that has grown hugely in importancerecently (deMenocal, 2001; Catto and Catto, 2004).

In many parts of the world, the effects of climate andsea-level changes on prehistoric societal evolution are wellunderstood and widely accepted. Classic studies include theeffects of drought on the collapse of various Asian civi-lizations during the early and middle Holocene (Rothman,2001; Weiss, 2000), the effects of increasing droughtincidence and severity on societies in central and southwestNorth America during the last two millennia (Jones et al.,1999; Haug et al., 2003), and the global effects of generallycooler, more variable climates during the Little Ice Agebeginning about A.D. 1350 (Grove, 1988; Fagan, 2000).Sea-level changes have also influenced prehistoric culturalevolution, both by periodically drowning areas valued byhumans, as in eastern China (Zhang et al., 2005), and byradically altering the nature and degree of productivity ofnearshore ecosystems thereby impacting the lifestyles ofdependent humans, as in many Pacific Islands (Esaka,1943; Amesbury, 1999). All such studies, fuelled by anincrease in the amount and precision of relevant paleocli-mate information in various parts of the world, is leading toa revival of the idea of environmental determinism.

This paper describes the nature of Pacific Basin societiesduring the Medieval Warm Period and Little Ice Age,followed by a number of examples illustrating the contrastbetween the ‘times of plenty’ and ‘times of less.’

Medieval Warm Period (Little Climatic Optimum)

During the Medieval Warm Period (approximately A.D.

700–1250) there is ample evidence that human societiesin most parts of the Pacific Basin increased in sociopoliticalcomplexity, manifested as development of social inequalityand stratification, the construction of monumental architec-ture, and the development of long-distance trade networks(Kirch, 1984, 1990; Ames, 1995). Most ideas about whyhuman societies increased in complexity at this time refer tolargely unverifiable notions such as increasing populationsand the arbitrary attainment of particular sociopoliticalthresholds (Kirch, 1984; Rechtman, 1992; Clark andHerdrich, 1993). A verifiable model has been suggestedthat involves the rising temperatures of the Medieval WarmPeriod increasing aridity across much of the region andforcing the amalgamation of smaller dispersed communitiesinto larger units (Fig. 1a, b) that could construct andmaintain water-conservatory food-production systems inways that smaller units could not. On account of their size,these larger units then became stratified and complex (Nunnand Britton, 2001; Nunn, 2003).

Increasing societal complexity was manifested in manyways, not simply by the oft-cited development of hierarchies

but also by the growth and maintenance of cooperativeenterprises (such as water-conservatory strategies for agricul-ture) and far-ranging ocean trading networks (Kirch, 1984;Nunn and Britton, 2001; Rolett, 2002). Among the formerare strategies for supplying water to water-needy communi-ties, such as the construction of canals (as in coastal Peru(Shimada, 2000)) and the irrigation of lowland crops (as onAneityum Island, Vanuatu (Spriggs, 1997)). Strategies forconserving available water, ranging from the establishmentof agricultural terraces in steep areas to conserve runoff,particularly in the tropical Pacific (Earle, 1980; Lucking,1984), to the construction of sunken food gardens, particu-larly along the arid fringe of South America (Rowe, 1969)and on Pacific Ocean atolls (Weisler, 1999), appear to datefrom the Medieval Warm Period in many places. Long-distance Pacific Ocean travel for trade and explorationreached its apogee during the later part of the MedievalWarm Period, as exemplified by the colonization of NewZealand about A.D. 1250–1300 from a tropical base at least2,900 km distant (Anderson, 1991; Hogg et al., 2003).

In most parts of the Pacific Basin during the MedievalWarm Period, human communities were concentrated incoastal areas, reflecting the critical role played by marinefoods in contemporary human livelihoods (Fig. 1a, b). Mostcommunities appear to have been able to gather/huntsufficient such food from nearshore areas. Most sedentarycoastal communities during the Medieval Warm Periodsupplemented marine foods with staples gathered/grown inlowland (coastal or valley-floor) areas or in upland areas thathad been modified to render them suitable for agriculture.

There are very few signs of intertribal conflict amongadjacent Pacific communities during the Medieval WarmPeriod. Reflecting this, most settlements were in openlocations rather than naturally defensible positions. Caveand rockshelter occupations were rare. Benign contactsamong both nearby and distant communities are suggestedby the appearance of artifacts (in particular) from one areain others. Examples include the shell of the pearl oyster(Pinctada margaritifera) from the Aitutaki Island lagoontraded throughout the southern Cook Islands (Walter,1990), ritual exchange between coastal and mountain(Andean) people in South America of the mollusc Spondy-lus princeps (Lumbreras, 1988), and obsidian circulation invarious parts of the Pacific Rim (Braswell and Glascock,2002; Kuzmin et al., 2002a, 2002b).

Little Ice Age

In most parts of the Pacific Basin, the Little Ice Age (A.D.1350–1800) saw a radical transformation in human lifestyles,expressed by conflict, by changing settlement patterns(Fig. 1c) and diets, and by a breakdown in the kinds of

Hum Ecol

interactions that had become commonplace during theMedieval Warm Period. It has been argued that the ultimatecause of this societal breakdown was environmental, foodresources in many areas having been significantly reduced bythe effects of cooling and sea-level fall (Nunn, 1999, 2000).This scenario is supported by considerable data, includingdetailed case studies from Fiji and Palau (Kumar et al., 2006;Masse et al., 2006). In parts of coastal Peru, it has beensuggested that massive flooding associated with a ‘mega ElNiño’ between A.D. 1300 and A.D. 1350 caused suchdisruption to the agricultural productivity of some commu-nities (such as the Chimú) that, rather than try to reestablishfarming, they sought to conquer more productive regions(such as the Lambayeque (Shimada, 2000)).

In many parts of the Pacific Basin, the open, commonlycoastal, settlements that existed during the Medieval WarmPeriod were abandoned during the Little Ice Age. On islands,where relocation beyond spheres of conflict was difficult,naturally defensible sites, particularly on hilltops or in caves,were established (Nunn, 2000; Green, 2002; Carson, 2003).Smaller, comparatively resource-poor, offshore islands wereoccupied for the first time around the start of the Little IceAge; examples come from Kapingamarangi Atoll in theMarshall Islands (Leach and Ward, 1981), the southernCook Islands (Walter, 1996), and the Line Islands of easternKiribati (di Piazza and Pearthree, 2001a). In many parts ofthe Pacific Rim, coastal settlements were abandoned duringthe early part of the Little Ice Age but often relocated inplaces beyond spheres of conflict, frequently where alter-natives to marine foods were plentiful (Paulsen, 1976;McGoodwin, 1992).

While many smaller Pacific Island communities did notstop consuming marine foods during the Little Ice Age,livelihoods based on inland, often upland, areas becamemore widespread at this time. Agricultural terraces in openareas were commonly abandoned; at the same time terraceswere established around the peripheries of fortified hilltopsettlements. The rise of cannibalism at this time can beexplained largely as an expression of conflict. Intraregional

and long-distance interactions between different groups ofpeople for the most part ended during the early part of theLittle Ice Age in many parts of this vast region (Fornander,1969; Irwin, 1992; Weisler, 1995, 1996; Luders, 1996).

It seems that the food resource base in many parts of thePacific Basin became more productive towards the end ofthe Little Ice Age. This was expressed by a lessening ofconflict—much warfare had become conventionalized bythe early nineteenth century (Nunn, 2000)—and thetentative movements of settlements back onto lowland andcoastal areas. Intraregional contacts began again in thoseareas where they had earlier ceased.

Examples

The examples below describe the nature of human societies inselected parts of the Pacific Basin during the Medieval WarmPeriod and Little Ice Age. The five case studies are intended toexemplify not only the common societal responses to the A.D.

1300 Event, as shown by the contrast between the times ofplenty (Medieval Warm Period) and the times of less (LittleIce Age), but also the variation in these responses dependingon the antecedent conditions. These include populationdensity, the breadth of available subsistence strategies, andpossibly the strength and synchronicity of the environmentalchanges in various parts of the Pacific Basin. The twogeneric examples (cave and rockshelter occupations, oceanvoyaging) are intended to show the Pacific-wide character ofresponses to the A.D. 1300 Event.

Case Study: Fiji

Although people living in Fiji during the Medieval WarmPeriod had a broader subsistence base covering largergeographical areas than those of earlier times, most remainedtethered to the coast. Except in a few areas, there are hardlyany truly inland (defined as >5 km from the coast) settlementsknown from Fiji prior to the A.D. 1300 Event. Most of those

Fig. 1 Changes in settlementpattern during the last1,200 years on many high Pa-cific Islands (after Nunn, 2003).Map a shows the situation at thestart of the Medieval WarmPeriod, perhaps about A.D. 900.Map b shows the situation at theend of the Medieval WarmPeriod, just prior to the A.D.

1300 Event about A.D. 1250.Map c shows the situation dur-ing the Little Ice Age, after theend of the A.D. 1300 Event aboutA.D. 1500.

Hum Ecol

appear to have been occupied only occasionally fordefensive reasons (such as the inland fort named Ulunikoroon Lakeba Island (Best, 1984; see Fig. 3)) or by groups intransit or on wild-food gathering expeditions, or by peoplewho exploited valley-floor resources, perhaps seasonally(such as the Sigatoka Valley (see Kumar et al., 2006)).

Oral traditions cannot generally reach back far enough intime to capture details of life in Fiji during the MedievalWarm Period and, owing to its comparatively low archaeo-logical visibility, the period has proven generally lesspopular for study than those earlier or later. Neverthelessthere is some information that supports the general model ofFiji communities living close to the coast during this period,particularly on the smaller islands of the group where feweroptions for subsistence were available than on larger islands.Two smaller islands are discussed here—Naigani in thecentral part of the group and Yadua in the north.

The island Naigani (1.9 km2) lies 11 km off the east coastof Viti Levu, the largest island in the group. About 800–900B.C., during the earliest (Lapita) phase of Fiji history, theisland was colonized by people who established a settlementat Matanamuani, on a tombolo adjoining what is believed tohave been one of the first broad fringing reefs to have

existed in Fiji waters (Fig. 2a; Best, 1981; Nunn, 2005).Although direct evidence is lacking, it is likely that Naiganisettlements remained exclusively coastal during the Medie-val Warm Period, having spread from Matanamuani to otherparts of the island’s coast as fringing reef developed thereunder the influence of falling late-Holocene sea level (Nunnand Peltier, 2001; Nunn, 2005). It is possible that peopleabandoned Naigani, notoriously drought-prone even today,for larger better-watered Viti Levu Island during theincreasingly arid Medieval Warm Period (Fig. 2b).

The island Yadua (13.6 km2) was colonized about thesame time as Naigani and they appear to have had a parallellast-millennium history. Well-preserved oral traditionssuggest that just prior to the A.D. 1300 Event the solesettlement on the island was on the west-facing Cukuvoucoastal flat (Nunn et al., 2005).

Throughout Fiji, the start of the Little Ice Age is markedby the Vuda Phase of settlement, during which warfarecommenced and inland, upland settlements were estab-lished. There is some disagreement amongst archaeologistsas to the precise inception date of the Vuda Phase, althougha recent summary concluded that “the heaviest phase offortification in Fiji ... did not begin for another few

Fig. 2 Summary diagram to show the timing of major changes insettlement pattern on Naigani (Ramoli and Nunn, 2001) and thecourse of late Holocene sea-level change (after Nunn and Peltier,2001). a—Location of the Matanamuani (Lapita) settlement 800–900B.C. b—Inferred situation during Medieval Warm Period with possible

out-migration owing to aridity. c—Reported situation from oraltraditions during early Little Ice Age involving abandonment of thethree coastal settlements in favor of ones on adjoining mountain-tops.d—Early 20th century saw abandonment of mountain-top settlementsin favour of lowland Natokalau and later modern Navitilevu.

Hum Ecol

centuries [after A.D. 950] with widespread warfare continu-ing well into the latter half of the nineteenth century”(Burley and Clark, 2003, p. 244). On the islands ofsouthern and central Lau (eastern Fiji), the period c. A.D.

1450–1750 marked the wholesale movement of peopleinland and upslope (Fig. 3) while on Beqa Island, inlandfortified sites dating from the last 500 years “were found allover the island, on every available peak” (Crosby, 1988, p.260). In the flat swampy Rewa Delta of southeast Viti LevuIsland, open sites were replaced by fortified ring-ditch sitesafter about A.D. 1450 (Marshall et al., 2000).

On Naigani Island, coastal settlements were abandonedat the start of the Little Ice Age and at least three mountain-top settlements established (Fig. 2c). Oral traditions suggestthat the occupants of these settlements remained incontinuous conflict throughout the Little Ice Age, ceasingonly around A.D. 1800 when the lowland Natokalausettlement was established. This was abandoned about100 years later in favor of the Navitilevu settlement whichfaced the direction of the new colony’s capital (Fig. 2d).

On Yadua Island, as warfare affected the island, so theCukuvou settlement was abandoned and a number of

Fig. 3 Last-millennium settle-ment history for Lakeba Island(56 km2), eastern Fiji, as shownby locations and numbers ofsettlements in different periods(distinguished by pottery style)(after Best, 2002). Note thatduring the Medieval Warm Pe-riod, settlements are all near-coastal; their low number mightbe because they were large and/or have low archaeological vis-ibility. Alternatively, aridityduring the Medieval Warm Pe-riod may have led most peopleto abandon islands like Lakeba.During the A.D. 1300 Event,settlements begin moving inlandand upslope, a trend that is at amaximum duringthe Little Ice Age.

Fig. 4 Schematic view of sea level and potential land use areas on the island Kaua’i during the Medieval Warm Period (a—Wainiha Valley, b—Mana Plain) and Little Ice Age (c—Wainiha Valley, d—Mana Plain).

Hum Ecol

mountain-top settlements established. As the threat ofconflict diminished towards the end of the nineteenthcentury, so the people moved to the east-facing Denimanucoastal flat where they remain today (Nunn et al., 2005).

Detailed work in the Sigatoka Valley (southwest Viti LevuIsland) has confirmed this general picture (Field, 2004; Nunnand Kumar, 2004; Kumar et al., 2006). Most settlement inthis part of Viti Levu was along the coast during theMedieval Warm Period but shortly after the A.D. 1300 Event,people began moving up-valley and offshore in largenumbers. Inland settlements were established in defensivelocations. Dates for alluvial charcoals show a sharp increasein upland vegetation clearance at this time, as do basal datesfor the main period of sand dune formation at the rivermouth. Comparable evidence is available elsewhere in Fiji,particularly from smaller islands (Rechtman, 1992;Cochrane, 2004); data from Lakeba are shown in Fig. 3.

Case Study: Kaua’i Island, Hawaiian Islands

In terms of the material archaeological patterns detectable at aregional scale, the archaeological signature of the A.D. 1300Event is probably more evident on larger islands such asKaua’i in the Hawaiian Archipelago. Case studies of theWainiha Valley and the Mānā Plain on Kaua’i exemplifythe physiographic diversity of this island group. Although thetiming and magnitude of the changes in climate and sea levelassociated with the A.D. 1300 Event are not as precisely orrobustly documented as desirable, the projected modelaccords well with the quantity, spatial distribution, and chro-nological association of archaeological sites in the two areas.

The Wainiha case represents the circumstances of mostphysiographic and cultural settings in the HawaiianArchipelago. It includes a coastal plain with a slopingmountainous interior valley, as appears to typify most landareas in the region, and contains one of the largest andwettest valleys of Kaua‘i: an attractive location for humansettlement and traditional land use. Figure 4a shows aschematic view of sea level and potential land use areas inWainiha Valley during the Medieval Warm Period.

The rising sea level during the Medieval Warm Periodaccordingly raised the water table, so the coastal areas andlower valley in Wainiha supported vibrant plant growth at thistime. The remains of a residential structure at the beach wasdated to A.D. 1030–1400, indicating a coastal-orientedcommunity probably during the earlier portion of this timerange (Carson, 2003, 2004a). In the upper valley, excavationsrevealed an undated period of human access to the uplandsprior to the creation of the extensive agricultural complexesvisible on the surface today (Barrera 1984a, 1984b).

The Mānā Plain case is atypical of most settings in theHawaiian Archipelago, and it affords a key opportunity toconsider what was actually affecting the overall patterns in

more representative places. Prior to drainage and filling forthe sugar cane industry in the late nineteenth and earlytwentieth centuries, the Mānā Plain was for the most part alow-elevation coastal marshland, bordered by a ridge-likebeach dune along the shoreline. Figure 4b shows a schematicviews of sea-level position and potential land use areas inthe Mānā Plain during the Medieval Warm Period. Duringthis period the Mānā Plain was almost entirely inundatednear the coast, although a few small patches of dry usablehigh ground existed amidst the extensive marsh. An isolatedfire pit was dated to A.D. 1150–1310 (Sweeney, 1994),apparently related to a rare temporary camp in the MānāPlain toward the end of the Medieval Warm Period.

After about A.D. 1300–1400, the residential occupation atthe beach in Wainiha was replaced by an extended period ofrepeated temporary camps and other short-term uses, whilethe focus of long-term residence moved somewhat fartherinland (Fig. 4c; Carson, 2003, 2004a). This remarkablechange in land use coincided with the proposed lower sealevel of the Little Ice Age, when a corresponding drop inthe water table may have reduced the productivity ofcoastal lands and also adversely affected the bounty of theoffshore reef ecosystem.

During the Little Ice Age, the projected lower sea levellikely reduced the marshy conditions of the Mānā Plain,and the high beach dune that developed could support moreregular, sustained, and repeated access to both marine andmarshland resources (Fig. 4d; Carson, 2004b).

Elsewhere in the Hawaiian Islands, coastal settlementswere also abandoned for island interiors during the Little IceAge. In the synthesis by Hommon (1986), the period A.D.

1400–1600 is marked by profound cultural and lifestyletransformation, and the large-scale expansion of inlandsettlement. Conflict and food scarcity led to the significantoccupation of marginal areas, such as the ‘Ewa Plain, a hotdry emerged limestone reef with numerous sinkholes onO’ahu Island, after A.D. 1300 (Athens et al., 2002) and theinhospitable water-limited Kahikinui area of Mau’i Islandaround A.D. 1500 (Stock et al., 2003).

The Wainiha example is a good illustration of thecontrast between the domination of coastal settlementduring the Medieval Warm Period and the inland settlementduring the Little Ice Age. The atypical Mānā Plain showsthe reverse, consistent with a fall of sea level.

Case Study: Easter Island

Easter Island (Rapa Nui) is a 142 km2 volcanic island some1,770 km from Pitcairn Island, the nearest neighbor to thewest, and 3,590 km from the South American coast. EasterIsland experiences highly variable monthly and annualrainfall and nearly constant winds. The soils are shallowand well drained. Fresh water is available in small,

Hum Ecol

dispersed springs and derives largely from intense bursts ofrainfall. In part due to cool sea temperatures and anextratropical insolation pattern, the marine biota isimpoverished (DiSalvo and Randall, 1993). The fewbeaches are windy and, like the predominantly rockyshorelines, are constantly exposed to high surf sincebuffering coral reefs are absent.

Easter Island was settled well after most of Polynesia(Spriggs and Anderson, 1993). Although different chronol-ogies of the island’s history have been proposed, one of themost compelling is that of Stevenson (1997). In this, theearliest occupation (Phase I) began by A.D. 800 in threeterrestrial zones: shoreward ritual stone platforms werebuilt; inland of the platforms were houses paired with earthovens; and inland of the houses and earth ovens weretemporary sites used as bases for resource collection. PhaseII covers the period A.D. 1100–1425, and does not includeany obvious effects attributable to the A.D. 1300 Event. Infact, activities typical of the mid- to late Medieval WarmPeriod elsewhere are suggested: more platforms near thecoast, and more houses and earth ovens inland of theplatforms. A significant change was that houses replacedthe temporary occupations farther inland, suggesting morepermanent occupation of the island’s interior.

The ‘times of plenty’ associated with the MedievalWarm Period are probably manifested by Phases I and II.The approximate coincidence of human arrival in Hawaii,New Zealand, and Easter Island in an accelerated popula-tion radiation into the eastern Pacific (Irwin, 1990) suggestsfavorable conditions for settlement at this time (Dickinson,2003); it is likely that some groups reached Central andSouth America (Anderson, 2003). Crops brought to EasterIsland from already-occupied islands included yams,coconut, sugarcane, taro, bananas, gourds, Cordyline, andmulberry (Métraux, 1971). Sweet potatoes—of New Worldorigin—were being grown on Easter Island by A.D. 1200 asindicated by charred tuber fragments from this crop inhearth deposits (Pearthree, 2003).

While initial settlement on many Pacific Islands wascoastal, people on Easter Island have generally avoided theshoreline for residence because of its inhospitable character.Inland residence near agricultural sites is also logical giventhe islanders’ spatially dispersed agricultural productionsystem. When crops are of relatively low caloric value(starches), it is less costly in terms of energy expenditurefor consumers to live near the crops than to move the cropsto consumers residing elsewhere. Conversely, fish is a highquality food (in calories and nutrition), and its capture andtransport from the sea to inland consumers is cost effective.

Easter Island’s culture probably evolved both throughcultural contacts and in isolation. The island is a ‘sink’habitat (Pulliam, 1996) where immigrants are necessary tomaintain a population over time. Cultural arrangements

among occupants of sources and sinks that ensure the flowof immigrants and other subsidies enable sink populationsto be sustained at an ‘unnatural’ density, higher than couldbe sustained without these arrangements. Through technicalimprovements over time, an island sink habitat might beconverted into a ‘pseudo-sink,’ capable of supporting alower density population in the absence of immigrants andother subsidies. This may have happened at Easter Island asclimatic conditions deteriorated after the A.D. 1300 Eventand interisland connections waned and finally ceasedaltogether. Similar scenarios have been suggested for theremote Pacific islands of the Pitcairn group (Weisler, 1996)and the Line Islands (Di Piazza and Pearthree, 2001b). Thedevelopment of such pseudo-sink islands may explain whytheir populations appeared initially to weather the environ-mental crisis associated with the A.D. 1300 Event, althoughsuccumbing eventually.

Archaeological data support the idea that cooler temper-atures and droughts during the Little Ice Age led to moreintensive agriculture on Easter Island, reflected in numerousinland agricultural and residential features. Multiple occu-pation layers may indicate residential mobility or increaseddispersal of a similar-sized population across the landscaperather than intrinsic growth. Figure 5 shows the frequencyof house occupation episodes in an inland zone in the southof the island from the period A.D. 900–1850. This area, themost favorable for agriculture on Easter Island, appears tohave been more intensively occupied towards the end of theMedieval Warm Period, with a slight increase about A.D.

1300 and a marked rise beginning about A.D. 1425 andcontinuing until about A.D. 1600, after which there is amarked decline.

The high-frequency inland occupation period in Fig. 5corresponds to Stevenson’s Phase III when “the Easter Islandpopulation reached its most complex level of hierarchicalorganization“ that may indicate ”a pattern of regionalstability where alliance building has resulted in a long-lasting and mutually satisfying arrangement” (Stevenson,1997, p. 11). A more plausible interpretation, consistent withthe evidence throughout the Pacific Islands and with that oflikely climate drivers, is that the considerable human energyinvestment in diverse farming and residential features, aswell as in numerous ritual stonework features (giant statues,elaborate burial platforms, ‘elite houses’ near the platforms)signals increasingly intense intergroup competition forlegitimate access to dwindling agricultural production. Inthis interpretation, by Phase III, the people of Easter Islandwere making the best of a worsening situation. Theycontrolled available runoff and minimized moisture loss byconstructing water diversion facilities, stone-walled agricul-tural pits, and stone mulches, for example, while participat-ing in social group-legitimizing rituals in the highlyconventionalized stonework idiom for which the island is

Hum Ecol

renowned. People were probably not especially ‘satisfied’with these arrangements but rather more ‘anxious’ aboutwhat was happening to their island. And, as elsewhere in thePacific Islands, those anxieties increasingly found expressionin conflict; the island’s obsidian began being mined for themanufacture of mata’a (spearheads) only around A.D. 1300, atime also marked by increasing occupation of cave androckshelter sites (Bahn and Flenley, 1992).

On the basis of shifts in climate ratios derived from theRano Aroi paleosediment core, Hunter-Anderson (1998)suggested that with the onset of the Little Ice Age there wasgreater climatic variability. Changes in woodland pollenand charcoal frequencies in dated archaeological assemb-lages indicate that by A.D. 1425, the endemic palm(Paschalococos dispersa) was in rapid decline, and theisland’s woody flora became increasingly impoverishedtowards the end of the sixteenth century or start of theseventeenth century, in contrast to its more diverse andmesic state in earlier centuries (Orliac, 2003).

The forest declined but palms and other ligneous speciesremained in places, despite deteriorating climate. In a studyof Easter Island’s ‘primeval’ soils, Mann et al. (2003) foundevidence for widespread slope wash beginning after A.D.

1200 (Fig. 6), and a new coring record at Rano Rarakushows a similar trend. This is consistent with increasedrainfall and/or with people changing land use in inland,hitherto largely unutilized, parts of the island. The base ofan erosional disconformity overlying palm root casts on thewestern distal slopes of Rano (crater) Kau was dated toabout A.D. 1620 (330±45 B.P.) indicating the persistence ofEaster Island’s palms despite human presence. Slope washand erosional deposition ceased as a major landscapealtering factor after about A.D. 1600 (Fig. 6).

The contracted settlement pattern of Phase IVcorresponds to the cessation of severe erosion and the

marked decline in inland usage after A.D. 1600 (Fig. 6).After about A.D. 1650 there was a major change in cookingfuel from primarily ligneous flora to the use of herbaceousforms, including native grass rhizomes and stalks (Orliacand Orliac, 1998). This is interpreted as a consequenceof a rapid spread of grasses over the island anddecimation of trees from a sudden and severe drought(perhaps also fire) followed by “famine, high mortality

Fig. 5 Frequency of rectangularhouse occupations in a portionof Easter Island’s southern inte-rior (after Stevenson, 1997). Themain period of sea-level fallduring the A.D. 1300 Event isshown as a shaded bar (seeFig. 8). Note that the start of thiscoincides with the first burst ofinland colonization, implyingthat the process may have beendriven by the associated coastal-resource depletion and/or water-table fall. The end of the A.D.

1300 Event, as shown, coincideswith the start of rapid inlandcolonization of this area, alsoimplying that this process mayhave been a result of associatednon-human drivers.

Fig. 6 Radiocarbon dates for soil charcoals (black bars, 2 sigmaerrors) plotted against cultural-historical events on Easter Island (afterMann et al., 2003). It is likely that the soil-charcoal dates representtimes of increased precipitation, possibly increased storminess,together with the effects of people altering the vegetation inpreviously unoccupied areas. The main period of sea-level fall duringthe A.D. 1300 Event is shown as a shaded bar (see Fig. 9).

Hum Ecol

and profound social disarray” (Orliac and Orliac, 1998,p. 132). Such changes are similar to those that occurredaround the same time on most other tropical PacificIslands. This observation demands a regional rather than alocal driving mechanism, and the A.D. 1300 Event appearsthe likeliest candidate.

Case Study: New Zealand

It is generally accepted that New Zealand was colonized byhumans around A.D. 1250–1300 (700–650 cal yr B.P.), oneof the latest dates for Pacific Islands (Anderson, 1991;McGlone and Wilmshurst, 1999; Hogg et al., 2003). Thecolonization of New Zealand was challenging in manyways, not least because it involved extraordinarily long-distance (?2,900 km) voyages and also the occupation oftemperate islands by people accustomed to tropical living.Some of the lifestyle components that the colonists mayhave thought to transfer to New Zealand—such as pigs andchickens, coconuts and breadfruit—did not succeed thereand most of the earliest communities that flourished appearto have done so largely on the basis of foods that weregathered or hunted (Anderson, 1997). A widespreaddependence on coastal foods, typically shellfish andnearshore fish species, is shown by the fact that for thefirst hundred years or so of settlement, “small communitieswere scattered along the coasts of both islands” of NewZealand (Grant, 1994, p. 184). One successful introductionwas the sweet potato (Ipomaea batatas), known as kumarain New Zealand, from the tropics, to which it is native, bysome of the islands’ earliest colonizers (McGlone et al.,1994; Langdon, 2001). Kumara was grown successfully inmany parts of North Island during the first century or so ofand became the single most important food crop for itsinhabitants.

The cooling marking the A.D. 1300 Event had a massiveeffect on horticulture in New Zealand because the wide-spread staple kumara, which was growing close to its lowertemperature tolerance limit in temperate New Zealand duringthe Medieval Warm Period, could not survive in the coolertemperatures that persisted after the A.D. 1300 Event. Thissaw a profound shift in lifestyles for many people from onesbased at least in part on horticulture to ones based solely onwild foods, principally bracken fernroot (Pteridium esculen-tum) (McGlone et al., 1994). Cooling may also have affectedthe ecosystems on which the herbivorous moa—the flight-less birds endemic to New Zealand—depended, althoughhumans appear largely responsible for their decline andextinction around A.D. 1550 (Anderson, 1989). Forestclearance using fire became widespread about the same time(McGlone, 1983), probably largely to facilitate access towild foods. Sea-level fall during the A.D. 1300 Event mayhave brought about significant losses of food resources

(especially shellfish) in nearshore areas (Rowland, 1976;Grant, 1994; Szabó, 2001).

The start of the Little Ice Age in New Zealand saw ashift from the coastal lowland settlements established by theislands’ colonizers to inland areas and upland fortifiedsettlements (named pa). Analysis shows that pa beganbeing built apparently simultaneously in every part of NewZealand about A.D. 1500 (Schmidt, 1995) although this datehas been challenged by Kirch (2000) who suggests that adate of A.D. 1300, originally suggested by Groube (1971), ismore realistic. At the time of European settlement, pa wereubiquitous in most of New Zealand, and “the earliestEuropean explorers found many tribal groups in a state ofconstant preparedness for warfare” (McGlone et al., 1994,p. 157). The establishment of pa settlements was associatedwith lifestyle changes and much thought has been given tothe cause-and-effect relationships (Davidson, 1987;McGlone et al., 1994). In some places the establishmentof pa was matched by piecemeal inland forest clearancecarried out by discrete fires up valley sides; an exampledated to A.D. 1201–1470 comes from the CoromandelPeninsula (Byrami et al., 2002).

Most authorities link the appearance of pa settlementswith the rise of conflict and attribute this to food shortagesassumed to be primarily a result of population increase. Aselsewhere in the Pacific Basin, this chain of deduction,while mostly uncontroversial, overlooks the regionalsynchronicity of both potential climatic drivers and societalresponses. As elsewhere, it favors population increase, notbecause of any empirical data, but solely because noalternative driver could be conceived. Yet pa haveequivalents throughout the Pacific Islands (Nunn, 2000)and are argued here as having been the stock societalresponse to conflict arising from food shortages on islandsof particular types and sizes with population densitiesabove a certain threshold (see Fig. 9b).

Case Study: Southern Queensland, East Australia

Southern Queensland is one of the most intensively studiedarchaeological provinces in Australia. Over the last15 years, general syntheses of regional archaeologicalpatterns have emerged that emphasize the synchronicity ofsignificant increases in site numbers and use after A.D. 950,especially around A.D. 1250 (Ulm and Hall, 1996).

The intensively studied southern Curtis Coast, forexample, reveals a near continuous record of occupationfrom 4000 B.P. to the present. The major trend evident in allanalyses is the dramatic increase in the number of sitescreated, sites occupied, and radiocarbon dates representedover the last 1,000 years, and particularly the last 700 years(Fig. 7). The trend is not gradual, but rather implies adisjunction in the regional trajectory of occupation. This

Hum Ecol

pattern is quite distinct from that evident in the preceding3,000 years of occupation and suggests a dramatic reorder-ing of land-use patterns, with no precedent in the history ofoccupation of this region.

Before the late Holocene, populations appear to have beenhighly mobile and wide-ranging, with evidence along thesouthern Queensland coasts for only occasional foragingexpeditions there from base camps inland. After A.D. 450 thecoast assumed an important role in regional mobilitystrategies, especially from around A.D. 1250. From thistime, changes in the quantity and composition of shellfishassemblages, localization in stone raw material sourcing,and discard of cultural materials together point to relativelymore intensive forms of land use than previous periods.

Archaeological investigations in other parts of southernQueensland have revealed similar patterns, in what Lourandosdescribed as “a more specialised and broad-based coastalemphasis in the economy of the most recent phase” (1997: p.161). Overall, these patterns point to a restructuring of land-

use from 1,000 years ago towards more systematic andpermanent coastal occupation with relatively low levels ofresidential mobility. These changes commenced around A.D.

750 in Moreton Bay (Ulm and Hall, 1996) and about A.D.

1050 for the Great Sandy Region (McNiven, 1999). To thenorth, increases in diet breadth in the Whitsunday Islandsfrom 600 years ago were associated with the emergence ofspecialized marine economies (Barker, 2004). The KeppelIslands were permanently occupied beginning only about700 years ago (Rowland, 1996).

In the last 1,000 years regional populations appear tohave been more territorially bounded or closed as denotedby marked localization of resource use and a dramaticincrease in the scale of coastal occupation. A recent greaterformalization of social relations is also suggested by theproduction and exchange of prestige items such as edge-ground hatchets. McNiven (1999) has argued that the recentintensive phase of occupation of coastal landscapesprovided economic, social, and political opportunities for

Fig. 7 Summed probability plot of all southern Curtis Coastcalibrated radiocarbon ages (n=56) for the establishment of humansettlement normalised to a maximum of one (see Ulm, 2004 for dates).The summed probability distribution represents the probability thatindependent events A or B occurred at a particular time and cantherefore be used as a proxy for the probability of occupation within aparticular period. Periods of low regional site use are common prior to1,000 years ago. Increasing amplitude of relative probability peaks

through time indicate a decrease in the spacing of dated occupationevents across the southern Curtis Coast after 1000 B.P. Within the last1,000 years marked increases are evident around 700 years ago. Themain period of sea-level fall during the A.D. 1300 Event is shown as ashaded bar (see Fig. 9). Note how this coincides with the period ofmost rapid rise of coastal settlement, implying that the effect ofcooling and sea-level fall on the island’s interior so depleted the foodresources that people moved to the coast.

Hum Ecol

the creation of new identity-conscious groups as part of ageneral restructuring of the social landscape.

It is plausible to suppose that these changes were drivenlargely by the cooling and sea-level fall of the A.D. 1300Event. Although precise pathways of cause and effect havenot been identified, and while it is also clear that they couldnot be the same as those on the more distant tropical PacificIslands, speculation is possible. For instance, in inland areasecosystem changes (associated with cooling during the A.D.

1300 Event) and water-table fall (driven by sea-level fall)may have made them drier and less productive in terms ofthe foods upon which their nomadic inhabitants depended,so that people congregated along the coast where foodresources were more abundant. The absence of conflict—such a common occurrence following the A.D. 1300 Eventin the distant Pacific Islands—is readily explained by thelow population densities relative to the foods availablealong the Queensland coast at this time. The existence oftensions associated with pressure on these food resourcesby adjoining sedentary communities could be inferred fromformulation of social relations manifested by prestige-itemexchange (see above).

The remarkable coincidence between the time of the abruptincrease in occupation of the southern Curtis Coast withthe earlier (largest and most prolonged) fall of sea level duringthe A.D. 1300 Event (Fig. 7) supports the idea that the lattercaused the former. Accepting this, the coincidence suggeststhat sea level fell significantly along the East Australiancoast in the period A.D. 1255–1398 (derived graphically fromFig. 9a) even though no such direct data have yet beenobtained for this period in this region.

Cave and Rockshelter Occupations

There is abundant evidence that most Pacific Basinsettlements during the Medieval Warm Period were coastal,or near-coastal in the sense that their occupants maderoutine use of nearby marine resources. Most such settle-ments were also open, in the sense that their locations werenot selected for their defensible attributes. During andfollowing the A.D. 1300 Event, many open and (near-)coastal settlements were abandoned in favor of smaller,often inland and upland, settlements in defensible locations(see Fig. 1 for general model; Nunn, 2000). The discussionin this section deliberately ignores the establishment offortified hilltop sites, for which a link with the A.D. 1300Event has been amply demonstrated (Nunn, 2000; Kumar etal., 2006; Masse et al., 2006), and focuses instead on theoccupation of caves and rockshelters.

Caves and rockshelters accommodated fewer people thanthe open sites of the Medieval Warm Period, and aretypically in areas that could be readily defended. The naturaldefensive qualities of many such sites were commonly

enhanced by the construction of ditches, walls and otherinfrastructure intended to help a small number of defenderssuccessfully repel a larger group of attackers (Fig. 8).

The switch from open (near-) coastal sites to caves androckshelters is most marked on smaller Pacific Islands ratherthan the larger islands (such as those in New Zealand) andthe contiguous continental margins of the Pacific whereother responses to conflict between people occupying opensites—principally involving migration out of conflictspheres—were easier than occupying defensible natural siteswithin a particular conflict sphere. For this reason, all theexamples given below come from smaller Pacific Islandsrather than larger islands and the Pacific Rim.

Cave (henceforth including rockshelters) occupations area frequent feature of the later prehistory of many smallerislands in the tropical Pacific. On Easter Island, caveoccupation began only about A.D. 1300, in Hawaii about400 years after initial settlement, and in the Marquesasabout 1,200 years after first settlement (Bahn and Flenley,1992). There are likely to have been many reasons for caveoccupation, depending on the nature of the caves available,the alternative settlement locations, and any environmentaland societal reasons for favoring (particular) caves. Yet it isclear that on many, mostly smaller, islands, cave occupa-tions increased sharply following the A.D. 1300 Event. Atthis time, many upland caves in defensible locations wereoccupied and fortified against attack. The same is true ofsome (near-) coastal caves although these were commonlyoccupied during the early Little Ice Age as temporaryshelters for marine food collection and processing.

The association of conflict with cave (and offshore islet)occupations on Easter Island is manifest from the deposits

Fig. 8 View of the entrance to the main cave at Tatuba, interior VitiLevu Island, Fiji. During the Little Ice Age, when the cave was adefensive site, the stone wall at the front of the cave extended fartherto allow limited access to the cave’s interior that could be easilyguarded. This mountaintop cave site is surrounded by stone walls andditches and was also used for defensive purposes during the colonialsuppression of the island’s interior people (Ravuvu, 1987). Photo byPatrick Nunn.

Hum Ecol

within the caves. Compared to the open settlements thatpreceded them, the occupants of the caves were primarilygathering and hunting wild foods, and absorbed in fashion-ing spear-points (mata’a) from the island’s obsidian. Caveoccupations and the exploitation of the island’s obsidian

sources both began about A.D. 1300 (see Fig. 9b) andcontinued until about A.D. 1650 (Bahn and Flenley, 1992).

On many islands following the A.D. 1300 Event, peoplemoved from open coast sites to cave sites along the sides ofnearby valleys. Initial rockshelter occupation in the Ana-

Fig. 9 Links between well-dat-ed chronologies of culturechange in the Pacific Basin andthe climate and sea-levelchanges of the A.D. 1300 Event.The shaded bar represents themain period of sea-level fallduring the A.D. 1300 Event inthe tropical Pacific, dated bygraphical fit in A to A.D. 1255–1398. a. Changes in tempera-ture, sea level and El Niñofrequency (Nunn, 2000). b.Chronology of colonization, firstestablishment of pa sites andmoa extinction in New Zealand(Hogg et al., 2003; Kirch, 2000;Anderson, 1989), with the earli-est dates for establishment ofnew hilltop/fortified sites in theSigatoka Valley, Fiji (afterKumar et al., 2006), date for asimilar site from Chuuk(Takayama and Intoh, 1978),hillfort establishment dates forpart of New Georgia, SolomonIslands (Sheppard et al., 2000),the commencement of caveoccupations and weapon manu-facture on Easter Island (Bahnand Flenley, 1992), and initialcave and valley-head occupa-tions on islands in French Poly-nesia (Orliac, 1997; Anderson etal., 2003; Kahn, 2003) for com-parison. c. Chronologies of Pa-cific Island ocean interactionfrom New Zealand (Anderson,1991; Buck, 1949), East Poly-nesia (Fornander, 1969),Vanuatu–Tonga (Luders, 1996),Fiji–Tonga (Reid, 1977), Hen-derson Island (Weisler, 1995,1996), eastern Kiribati (Kiriti-mati, Manra and Tabuaeranislands—Di Piazza and Pear-three, 2001b), Mangareva-Mar-quesas (Weisler and Green,2001), and Chatham Islands(interaction with New Zealand;Sutton, 1980). Note that in thelight of the late date for coloni-zation of New Zealand obtainedby Hogg et al. (2003), the earlierdate for the Chatham Islandsmust be doubted.

Hum Ecol

hulu Valley on O’ahu Island in Hawaii dates from about A.D. 1350 (Kirch, 1989). Other examples of late intensive useof caves in Hawaii include Maunalua in southeast O’ahu(Thomas, 1995) and Kaupikiawa along the rugged Kalau-papa Peninsula of Moloka’i (Kirch et al., 2003).

Good examples also come from French Polynesia. OnTahiti coastal settlements were abandoned in favor of valley-side cave occupations by at least A.D. 1370 (Orliac, 1997). OnMangareva Island (Gambier Islands, French Polynesia),where initial coastal settlement dates from at least A.D.

1160, a cave occupation up the Atiaoa Valley began in A.D.

1280–1300, while another in the Gatavake Valley began noearlier than A.D. 1660 (Anderson et al., 2003). In the‘Opunohu Valley on Mo’orea Island (Society Islands, FrenchPolynesia), inland, valley-side house platforms were estab-lished by the thirteenth century (Green, 1996); a more recentstudy dated some to no earlier than A.D. 1415 (Kahn, 2003).

Other good examples of cave sites come from the FijiIslands although most have not been dated securely, informa-tion about their human occupation during the Little Ice Agecoming from oral traditions. One exception is the caveQaranilaca on the eastern Fiji island Vanuabalavu, the humanuse of which intensified after the A.D. 1300 Event andinvolved assembling (and preparation?) of marine foodsthen carried to nearby mountain-top settlements for con-sumption (Thomas et al., 2004). Another exception is fromWaya Island in the west of the group, where the use of thecave Qaranicagi increased markedly between A.D. 1480–1640 (Cochrane, 2004). Caves in the Sigatoka Valley on VitiLevu Island were popular sites for occupation following theA.D. 1300 Event, some like Tatuba, shown in Fig. 8, beingelaborately fortified (Field, 2004; Kumar et al., 2006).

Ocean Voyaging

One of the major features of the Medieval Warm Period inthe Pacific Ocean is the comparative frequency of routineocean travel, both within and between island groups. Mostobservers regard the human colonization (not discovery) ofmost Pacific Island groups as having been preceded bypioneer voyagers who returned to their source to report theexistence of the distant lands, how to reach them, and whatkinds of environments they contained. For example, it isargued that the colonization of New Zealand from tropicalbases perhaps 2,900 km away involved deliberate voyagesof colonization with the migrants well prepared for theconditions they would encounter (McGlone et al., 1994).Yet two-way voyaging between New Zealand and thetropical islands of eastern Polynesia came to an abrupt haltabout A.D. 1350. This situation is paralleled throughout thePacific Islands, manifested primarily as discontinuedmaterial and linguistic exchange (Fornander, 1969; Lewis,1972; Irwin, 1992; Nunn, 2000).

The precise dating of the cessation of interislandcommunication is difficult to achieve. In a study of isolatedHenderson Island in the southeast Pacific, it was argued thatA.D. 1500 marked a change from imported artifacts to “analmost exclusive reliance on locally available raw materi-als,” implying the end of interisland contact (Weisler, 1996,p. 622). The island was finally abandoned about 200 yearslater. A similar picture has emerged for Easter Island withthe end of interisland contact undermining the culturalmomentum represented by the carving and erection of thegiant statues (moai) during the Medieval Warm Period andcontributing to a profound cultural transformation associat-ed with the A.D. 1300 Event (see above). Interaction withinthe Marquesas archipelago of French Polynesia, representedby the exchange of basalt and phonolite for tool manufac-ture, ended about A.D. 1450 (Rolett, 2002).

On some islands, the end of specialized long-distance tradecan be linked to the A.D. 1300 Event. A good example is theTonga–Vanuatu kava (Piper methysticum) trade that endedabout A.D. 1447 (Luders, 1996). The end of two-wayvoyaging between New Zealand and Eastern Polynesiaended about A.D. 1350, a date acquired from genealogies(Buck, 1949) yet one that agrees with cruder estimates. Onthe Chatham Islands, no further people arrived (in prehistorictimes) after A.D. 1400 (Sutton, 1980). A general synthesis forEast Polynesia (including Hawaii and New Zealand)concluded that long-distance voyaging ended abruptly atthe end of the fourteenth century (Fornander, 1969).

A selection of regional and sub-regional interactionsinvolving or representing ocean voyaging are shown inFig. 9c. It can be seen how long-distance interaction cameto an end within the broadly-defined A.D. 1300 Event whilesubregional interactions generally ceased within 100 yearsor so of its end. This can be explained by the effects ofchanged climate disrupting the weather and ocean con-ditions to which long-distance voyagers were accustomed.The slight continuation of subregional interaction mayreflect the continuing success of shorter voyages as wellas the imperatives of maintaining subregional networks forcultural reasons, including exchange, and survival. If theVanuatu-Tonga kava trade, representing ocean voyages ofabout 1,700 km, and the subregional interactions involvingHenderson Island and the Chatham Islands, representingvoyages of about 200–650 km and 860 km respectively,ceased because of climate change, then this could bear outthe idea that the A.D. 1300 Event was associated with aperiod of storminess that continued into the Little Ice Age.

The lack of renewal of significant Pacific-wide prehis-toric ocean voyaging after about A.D. 1500 is interpreted asa result of (a) the conflict that pervaded many Pacificsocieties at the time, (b) the establishment of upland/inland/offshore settlements and less marine-dependent lifestyles,and (c) a loss of seafaring and navigational expertise. The

Hum Ecol

renewal of Fiji-Tonga interaction around A.D. 1600, repre-sented here by the genealogical documentation of the firstmarriage between the eldest sister of the Tu’i Tonga(paramount chief of Tonga) and a chief of the eastern FijiIsland Lakeba (Reid, 1977), may have signalled the pointduring the Little Ice Age when both climatic and societalconditions became conducive to resumption of theseregular ∼425 km voyages that continued for the rest ofthe Little Ice Age resulting in political dominance ofTongans in many parts of the Fiji group.

These kinds of data have the potential to illuminate chro-nologies of the climate and sea-level changes that ultimatelydrove them. Yet before they can be used for this purpose, it isnecessary-as for many other cultural data—to identifyplausible cause-effect pathways and estimate their durations.

Conclusions

The last millennium in the Pacific Basin was characterizedby a sharp contrast between times of plenty and times ofless, largely a consequence of the influences of climate andsea-level change on human societies against the backdropof environments of varying sensitivity to change. Theunavoidable implication, similar to many recent studies ofother parts of the world (deMenocal, 2001; Berglund, 2003;Catto and Catto, 2004; Diamond, 2005), is that environ-mental change does appear to have been a significant causeof cultural change, and that the paradigm of environmentaldeterminism, long rejected, is in need of reevaluation.

There is great potential for understanding the sequenceand the chronology of the assumed environmental driversof the A.D. 1300 Event from studying the sequence andchronology of its assumed societal responses. At this stage,key cultural data can only confirm the general pattern ofsea-level fall and climatic deterioration linked to coolingduring the A.D. 1300 Event. Future work might focus ongathering more cultural data in support of the general modelbut also using those data to detect regional variations intiming and magnitude of climate and sea-level change.

Acknowledgements Unpublished oral traditions from Yadua Island(Fiji) were collected from Maikeli Rasese by Elia Nakoro. Unpub-lished oral traditions from Naigani Island (Fiji) were collected byPatrick Nunn. The U.S. Navy is acknowledged for grantingpermission to refer to data from archaeological work in the PacificMissile Range Facility at Barking Sands, Kaua‘i Island, Hawaii.

References

Ames, K. M. (1995). Chiefly power and household production on thenorthwest coast. In Price, T. D., and Feinman, G. M. (eds.),Foundations of Social Inequality, Plenum, New York, pp.155–187.

Amesbury, J. R. (1999). Changes in Species Composition ofArchaeological Marine Shell Assemblages in Guam. Micronesica31: 347–366.

Anderson, A. (1989). Prodigious Birds: Moa and Moa-Hunting inPrehistoric New Zealand. Cambridge University Press, Cam-bridge, UK.

Anderson, A. (1991). The Chronology of Colonisation in NewZealand. Antiquity 65: 767–795.

Anderson, A. (1997). Prehistoric Polynesian impact on the NewZealand environment: Te Whenua Hou. In Kirch, P. V., and Hunt,T. L. (eds.), Historical Ecology in the Pacific Islands. YaleUniversity Press, New Haven, CT, pp. 147–165.

Anderson, A. (2003). Initial human dispersal in remote Oceania:pattern and explanation. In Sand, C. (ed.), Pacific Archaeology:Assessments and Prospects (Proceedings of the InternationalConference for the 50th Anniversary of the First LapitaExcavation, Kone-Nouméa 2002). Services des Musées et duPatrimoine, Nouméa, New Caledonia, pp. 71–84.

Anderson, A., Conte, E., Kirch, P. V., and Weisler, M. (2003).Archaeological investigations of the Gambier Islands, FrenchPolynesia, 2001. In Sand, C. (ed.), Pacific Archaeology: Assess-ments and Prospects (Proceedings of the International Confer-ence for the 50th Anniversary of the First Lapita Excavation,Kone-Nouméa 2002). Services des Musées et du Patrimoine,Nouméa, New Caledonia, pp. 343–352.

Athens, J. S., Tuggle, H. D., Ward, J. V., and Welch, D. J. (2002).Avifaunal Extinctions, Vegetation Change, and Polynesian Im-pacts in Prehistoric Hawai’i. Archaeology in Oceania 37: 57–78.

Bahn, P. G., and Flenley, J. (1992). Easter Island, Earth Island.Thames and Hudson, London.

Barker, B. (2004). The Sea People: Late Holocene maritimespecialisation in the Whitsunday Islands, Central Queensland.Pandanus Books (Terra Australis 20), Canberra.

Barrera, W, Jr. (1984a). Upper Wainiha Valley, Kauai: ArchaeologicalSurvey. Report prepared for EDAW, Inc., Chiniago, Inc., HawaiiState Historic Preservation Division, Kapolei.

Barrera, W, Jr. (1984b). Wainiha Valley, Kauai: ArchaeologicalInvestigations. Report prepared for Orion Engineering, Chi-niago, Inc., Hawaii State Historic Preservation Division, Kapolei.

Berglund, B. E. (2003). Human Impact and Climate Changes—Synchronous Events and a Causal Link? Quaternary Interna-tional 105: 7–12.

Best, S. (1981). Excavations at Site VL 21/5 Naigani Island, Fiji: APreliminary Report. Department of Anthropology, University ofAuckland.

Best, S. B. (1984). Lakeba: The Prehistory of a Fijian island. Ph.Ddissertation, University of Auckland.

Best, S. (2002). Lapita: A View from the East. New ZealandArchaeological Association Monograph Vol 24, Auckland.

Braswell, G. E., and Glascock, M. D. (2002). The emergence ofmarket economies in the ancient Maya world: obsidian exchangein Terminal Classic Yucatán. In Glascock, M. D. (ed.),Geochemical Evidence for Long-distance Exchange. Bergin andGarvey, Westport, CT, pp. 33–52.

Buck, P. (Te Rangi Hiroa). (1949). The Coming of the Maori.Whitcombe and Tombes, Christchurch, New Zealand.

Burley, D. V., and Clark, J. T. (2003). The archaeology of Fiji/WestPolynesia in the Post-Lapita era. In Sand, C. (ed.), PacificArchaeology: Assessments and Prospects (Proceedings of theInternational Conference for the 50th Anniversary of the FirstLapita Excavation, Kone-Nouméa 2002). Services des Musées etdu Patrimoine, Nouméa, New Caledonia, pp. 235–254.

Byrami, M., Ogden, J., Horrocks, M., Deng, Y., Shane, P., and Palmer,J. (2002). A Palynological Study of Polynesian and EuropeanEffects on Vegetation in Coromandel, New Zealand, Showing theVariability Between Four Records from a Single Swamp. Journalof the Royal Society of New Zealand 32: 507–531.

Hum Ecol

Carson, M. T. (2003). Integrating Fragments of a Settlement Patternand Cultural Sequence in Wainiha Valley, Kaua‘i, HawaiianIslands. People and Culture in Oceania 19: 83–105.

Carson, M. T. (2004a). Resolving the Enigma of Early CoastalSettlement in the Hawaiian Islands: The Stratigraphic Se-quence of the Wainiha Beach Site in Kaua‘i. Geoarchaeology19: 98–118.

Carson, M. T. (2004b). Archaeological Monitoring in Support of theConstruction of New Test Facilities at Pacific Missile RangeFacility Barking Sands, Island of Kaua‘i, State of Hawai‘i.Report 200312 prepared by International Archaeological Re-search Institute, Inc., Honolulu under Contract N62742-97-D-3511 Task Order 0021 for Department of the Navy, PacificDivision, Naval Facilities Engineering Command, Pearl Harbor.

Catto, N., and Catto, G. (2004). Climate Change, Communities, andCivilizations: Driving Force, Supporting Player, or BackgroundNoise? Quaternary International 123–125: 7–10.

Clark, J., and Herdrich, J. (1993). Prehistoric Settlement System inEast Tutuila, American Samoa. Journal of the Polynesian Society102: 147–185.

Cochrane, E. E. (2004). Archaeological Investigations on WayaIsland: The 2001 University of Hawai’i Archaeological FieldSchool. Domodomo 17: 7–13.

Crosby, A. (1988). Beqa: Archaeology, Structure and History in Fiji. MAThesis, Department of Anthropology, University of Auckland.

Davidson, J. (1987). The children of Tu: warfare and fortifications. InWilson, J. (ed.), From the Beginning: The Archaeology of theMaori. Penguin, Auckland, New Zealand, pp. 109–124.

deMenocal, P. B. (2001). Cultural Responses to Climate ChangeDuring the Late Holocene. Science 292: 667–673.

Diamond, J. (2005). Collapse: How Societies Choose to Fail orSucceed. Viking, New York.

Dickinson, W. R. (2003). Impact of Mid-Holocene Hydro-isostaticHighstand in Regional Sea Level on Habitability of Islands inPacific Oceania. Journal of Coastal Research 19: 489–502.

Di Piazza, A., and Pearthree, E. (2001a). An island for Gardens, anIsland for Birds and Voyaging: A Settlement Pattern forKiritimati and Tabuaeran, Two “Mystery Islands” in the NorthernLines, Republic of Kiribati. Journal of the Polynesian Society110: 149–170.

Di Piazza, A., and Pearthree, E. (2001b). Voyaging and Basalt Exchangein the Phoenix and Line Archipelagoes: The Viewpoint from ThreeMystery Islands. Archaeology in Oceania 36: 146–152.

DiSalvo, L. H., and Randall, J. E. (1993). TheMarine Fauna of Rapanui,Past and Present. In Fischer, S. R. (ed.), Easter Island Studies,Contributions to the History of Rapanui in Memory of William T.Mulloy. Oxbow Monograph Vol 32, Oxford, pp. 16–23.

Earle, T. (1980). Prehistoric Irrigation in the Hawaiian Islands: AnEvaluation of Evolutionary Significance. Archaeology andPhysical Anthropology in Oceania 15: 1–28.

Esaka, T. (1943). Minami Kanto Shinsekki jidai kaizuka yori kantaruChusekisei ni okeru kaishin kaitai (Holocene sea transgressionsand regressions as seen in the Neolithic shellmounds of SouthKanto). Kodai Bunka 14(4). [in Japanese]

Fagan, B. (2000). The Little Ice Age: How Climate Made History,1300–1850. Basic Books, New York.

Field, J. S. (2004). Environmental and Climatic Considerations: AHypothesis for Conflict and the Emergence of Social Complexityin Fijian Prehistory. Journal of Anthropological Archaeology 23:79–99.

Fornander, A. (1969). An Account of the Polynesian Race, Its Originsand Migrations. Vol. 2, Tuttle, Tokyo.

Grant, P. J. (1994). Late Holocene histories of climate, geomorphologyand vegetation, and their effects on the first New Zealanders. InSutton, D. (ed.), The Origins of the First New Zealanders.Auckland University Press, Auckland, New Zealand, pp. 164–194.

Green, R. C. (1996). Settlement patterns and complex society in theWindward Society Islands. In Julien, M., Orliac, M., and Orliac,C. (eds.), Mémoire de Pierre, Mémoire d’Homme, Tradition etArchéologie en Océanie, Ouvrage Collectif en Hommage à J.Garanger. Publication de la Sorbonne, Paris, pp. 207–226.

Green, R. C. (2002). A retrospective view of settlement pattern studiesin Samoa. In Ladefoged, T. N., and Graves, M. W. (eds.), PacificLandscapes: Archaeological Approaches. Easter Island Books,Los Osos, CA, pp. 125–152.

Groube, R. C. (1971). The origin and development of earthworkfortifications in the Pacific. Studies in Oceanic Culture History,Pacific Anthropological Association Records 1: 133–164.

Grove, J. M. (1988). The Little Ice Age. Methuen, London.Haug, G. H., Günther, D., Peterson, L. C., Sigman, D. M., Hughen, K.

A., and Aeschlimann, B. (2003). Climate and the Collapse ofMaya Civilization. Science 299: 1731–1735.

Hogg, A. G., Higham, T. F. G., Lowe, D. J., Palmer, J. G., Reimer,P. J., and Newnham, R. M. (2003). A Wiggle-match Date forPolynesian Settlement of New Zealand. Antiquity 77: 116–125.

Hommon, R. J. (1986). Social evolution in ancient Hawaii. In Kirch,P. V. (ed.), Island Societies: Archaeological Approaches toEvolution and Transformation. Cambridge University Press,Cambridge, UK, pp. 55–68.

Hunter-Anderson, R. L. (1998). Human vs. climatic impacts at EasterIsland: Did the people really cut down all those trees? InStevenson, C. M., Lee, G., and Morin, F. J. (eds.), Easter Islandin Pacific Context, South Seas Symposium, Proceedings of theFourth International Conference on Easter Island and EastPolynesia, University of New Mexico. The Easter IslandFoundation, Los Osos, CA, pp. 85–99.

Irwin, G. (1990). Human Colonisation and Change in the RemotePacific. Current Anthropology 31: 90–94.

Irwin, G. (1992). The Prehistoric Exploration and Colonisation of thePacific. Cambridge University Press, Cambridge, UK.

Jones, T. L., Brown, G. M., Raab, M., McVickar, J. L., Spaulding, W.G., Kennett, D. J., York, A., and Walker, P. L. (1999).Environmental Imperatives Reconsidered: Demographic Crisesin Western North America during the Medieval ClimaticAnomaly. Current Anthropology 40: 137–170.

Kahn, J. (2003). Maohi social organization at the micro-scale:household archaeology in the ‘Opunohu Valley, Mo’orea, SocietyIslands French Polynesia. In Sand, C. (ed.), Pacific Archaeology:Assessments and Prospects (Proceedings of the InternationalConference for the 50th Anniversary of the First LapitaExcavation, Kone-Nouméa 2002). Services des Musées et duPatrimoine, Nouméa, New Caledonia, pp. 353–367.

Kirch, P. V. (1984). The Evolution of the Polynesian Chiefdoms.Cambridge University Press, Cambridge, UK.

Kirch, P. V. (1989). Prehistoric Hawaiian occupation in the AnahuluValley, O’ahu Island: Excavations in three inland rockshelters.Contributions of the University of California (Berkeley) Archae-ological Research Facility Vol. 47, Berkeley, CA.

Kirch, P. V. (1990). The Evolution of Sociopolitical Complexity inPrehistoric Hawaii: An Assessment of the ArchaeologicalEvidence. Journal of World Prehistory 4: 311–345.

Kirch, P. V. (2000). On the Road of the Winds: An ArchaeologicalHistory of the Pacific Islands before European Contact.University of California Press, Berkeley, CA.

Kirch, P. V., O’Day, S., Coil, J., Morgenstein, M., Kawelu, K., andMillerstrom, M. (2003). The Kaupikiawa Rockshelter, KalaupapaPeninsula, Moloka’i: New Investigations and Reinterpretation ofits Significance for Hawaiian Prehistory. People and Culture inOceania 19: 1–27.

Kumar, R., Nunn, P. D., Field, J. E., and de Biran, A. (2006). HumanResponses to Climate Change Around A.D. 1300: A Case Studyof the Sigatoka Valley, Viti Levu Island, Fiji. QuaternaryInternational 151: 133–143.

Hum Ecol

Kuzmin, Y. V., Glascock, M. D., and Sato, H. (2002a). Sources ofArchaeological Obsidian on Sakhalin Island (Russian Far East).Journal of Archaeological Science 29: 741–749.

Kuzmin, Y. V., Popov, V. K., Glascock, M. D., and Shackley, M. S.(2002b). Sources of Archaeological Volcanic Glass in thePrimorye (Maritime) Province, Russian Far East. Archaeometry44: 505–515.

Langdon, R. (2001). The Bamboo Raft as a Key to the Introduction ofthe Sweet Potato in Prehistoric Polynesia. Journal of PacificHistory 36: 51–76.

Leach, F., and Ward, G. (1981). Archaeology on KapingamarangiAtoll: a Polynesian outlier in the eastern Caroline Islands.Manuscript, Pacific Collection, The University of the SouthPacific Library, Suva, Fiji.

Lewis, D. (1972). We, the Navigators. Australian National UniversityPress, Canberra, Australia.

Lourandos, H. (1997). Continent of Hunter–Gatherers: New Perspec-tives in Australian Prehistory. Cambridge University Press,Cambridge, UK.

Lucking, L. J. (1984). An archaeological investigation of prehistoricPalauan terraces. PhD thesis, University ofMinnesota,Minneapolis.

Luders, D. (1996). Legend and History: Did the Vanuatu–Tonga KavaTrade Cease in A.D. 1447? Journal of the Polynesian Society105: 287–310.

Lumbreras, L. G. (1988). Childe y la tesis de la revolución urbana: laexperiencia central andina. In Manzanilla, L. (ed.), Coloquio V.Gordon Childe: Estudios sobre la revolución neolítica y larevolución urbana. Instituto de Investigaciones Antropológicas,UNAM, Mexico, pp. 349–366.

McGlone, M. S. (1983). Polynesian Deforestation of New Zealand: APreliminary Synthesis. Archaeology in Oceania 18: 11–25.

McGlone, M. S., and Wilmshurst, J. M. (1999). Dating Initial MaoriEnvironmental Impact in New Zealand. Quaternary International59: 5–16.

McGlone, M. S., Anderson, A. J., and Holdaway, R. N. (1994). Anecological approach to the Polynesian settlement of New Zealand.In Sutton, D. (ed.), The Origins of the First New Zealanders.Auckland University Press, Auckland, New Zealand, pp. 136–163.

McGoodwin, J. R. (1992). Human responses to weather-inducedcatastrophes in a west Mexican fishery. In Glantz, M. H. (ed.),Climate Variability, Climate Change, and Fisheries. CambridgeUniversity Press, Cambridge, UK, pp. 168–184.

McNiven, I. (1999). Fissioning and regionalisation: the social dimen-sions of changes in Aboriginal use of the Great Sandy Region,southeast Queensland. In Hall, J., and McNiven, I. (eds.),Australian Coastal Archaeology. Australian National University(Research Papers in Archaeology and Natural History 31,Archaeology and Natural History Publications, Research Schoolof Pacific and Asian Studies), Canberra, Australia, pp. 157–168.

Mann, D., Chase, J., Edwards, J., Beck, W., Reanier, R., and Mass, M.(2003). Prehistoric destruction of the primeval soils and vegetationof Easter Island. In Trancredi, J. T., and Loret, J. (eds.), EasterIsland, Scientific Exploration into the World’s EnvironmentalProblems in Microcosm. Kluwer, New York, pp. 133–153.

Marshall, Y., Crosby, A., Matararaba, S., and Wood, S. (2000).Sigatoka: The Shifting Sands of Fijian Prehistory. Oxbow Books(University of Southampton, Department of Archaeology,Monograph 1), Oxford.

Masse, W. B., Liston, J., Carucci, J., and Athens, J. S. (2006).Evaluating the Effects of Climate Change on Environment,Resource Depletion, and Culture in the Palau Islands betweenA.D. 1200 and 1600. Quaternary International 151: 106–132.

Métraux, A. (1971). Ethnology of Easter Island. Bulletin 160. BerniceP. Bishop Museum, Honolulu, HI.

Nunn, P. D. (1999). Environmental Change in the Pacific Basin:Chronologies, Causes, Consequences. Wiley, London.

Nunn, P. D. (2000). Environmental Catastrophe in the Pacific Islandsabout A.D. 1300. Geoarchaeology 15: 715–740.

Nunn, P. D. (2003). Nature–Society Interactions in the Pacific Islands.Geografiska Annaler 85 B: 219–229.

Nunn, P. D. (2005). Reconstructing Tropical Paleoshorelines UsingArchaeological Data: Examples from the Fiji Archipelago, South-west Pacific. Journal of Coastal Research, Special Issue 42: 15–25.

Nunn, P. D., and Britton, J. M. R. (2001). Human–EnvironmentRelationships in the Pacific Islands Around A.D. 1300. Environ-ment and History 7: 3–22.

Nunn, P. D., and Peltier, W. R. (2001). Far-Field Test of the ICE-4G (VM2)Model of Global Isostatic Response to Deglaciation: Empirical andTheoretical Holocene Sea-level Reconstructions for the Fiji Islands,Southwest Pacific. Quaternary Research 55: 203–214.

Nunn, P. D., and Kumar, R. (2004). Alluvial Charcoal in the SigatokaValley, Viti Levu Island, Fiji. Palaeogeography, Palaeoclimatol-ogy, Palaeoecology 214: 153–162.

Nunn, P. D., Matararaba, S., Ishimura, T., Kumar, R., and Nakoro, E.(2005). Reconstructing the Lapita-era Geography of NorthernFiji: A Newly-discovered Lapita Site on Yadua Island and itsImplications. New Zealand Journal of Archaeology 26: 41–55.

Orliac, M. (1997). Human occupation and environmental modifica-tions in the Papeno’o valley, Tahiti. In Kirch, P. V., and Hunt, T.L. (eds.), Historical Ecology in the Pacific Islands. YaleUniversity Press, New Haven, CT, pp. 200–229.

Orliac, C. (2003). Ligneus et palmiers de l’île de Pâques du XIème auXVIIème siècle de notre ère. In Orliac, C. (ed.), Archéologie enOcéanie Insulaire, Peuplement, Sociétés et Paysages. ÉditionsArtcom, Paris, pp. 184–199.

Orliac, C., and Orliac, M. (1998). The disappearance of Easter Island’sforest: Over-exploitation or climatic catastrophe? In Stevenson,C. M., Lee, G., and Morin, F. J. (eds.), Easter Island in PacificContext, South Seas Symposium, Proceedings of the FourthInternational Conference on Easter Island and East Polynesia,University of New Mexico. The Easter Island Foundation, LosOsos, CA, pp. 129–134.

Paulsen, A. C. (1976). Environment and Empire: Climatic Factors inPrehistoric Andean Culture Change.World Archaeology 8: 121–132.

Pearthree, E. (2003). Identification des restes carbonisés de plantes non-ligneuses découverts sur trois sites d’habitat à l’île de Pâques. InOrliac, C. (ed.), Archéologie en Océanie Insulaire, Peuplement,Sociétés et Paysages. Éditions Artcom, Paris, pp. 172–183.

Pulliam, H. R. (1996). Sources and sinks: Empirical evidence andpopulation consequences. In Olin, J., Rhodes, E., Chesser, R. K.,and Smith, M. H. (eds.), Population Dynamics in Ecological Spaceand Time. University of Chicago Press, Chicago, IL, pp. 45–70.

Ramoli, A., and Nunn, P. D. (2001). Naigani Island and Its HistoricalConnections with Ovalau and Moturiki Islands: ConvergencesBetween Legend and Fact. Domodomo 13: 19–28.

Ravuvu, A. D. (1987). The Fijian Ethos. Institute of Pacific Studies,The University of the South Pacific, Suva, Fiji.

Rechtman, R. B. (1992). The evolution of sociopolitical complexity inthe Fiji Islands. PhD thesis, University of California Los Angeles.

Reid, A. C. (1977). The Fruit of the Rewa: Oral Traditions and theGrowth of the Pre-Christian Lakeba State. Journal of PacificHistory 12: 1–24.

Rolett, B. V. (2002). Voyaging and Interaction in Ancient EastPolynesia. Asian Perspectives 41: 182–194.

Rothman, M. (ed). (2001). Uruk Mesopotamia and its Neighbors:Cross-Cultural Interactions in the Era of State Formation.Society for American Research, Santa Fe, NM.

Rowe, J. (1969). The Sunken Gardens of the Peruvian Coast.American Antiquity 34: 320–325.

Rowland, M. J. (1976). Cellana denticulata in Middens on theCoromandel Coast, NZ—Possibilities for a Temporal Horizon.Journal of the Royal Society of New Zealand 6: 1–15.

Hum Ecol

Rowland, M. J. (1996). Prehistoric archaeology of the Great BarrierReef Province—Retrospect and prospect. In Veth, P., andHiscock, P. (eds.), Archaeology of Northern Australia. Anthro-pology Museum, University of Queensland (Tempus 4), St Lucia,pp. 191–212.

Schmidt, M. (1995). The Commencement of Pa Construction in NewZealand Prehistory. Journal of the Polynesian Society 105: 441–460.

Sheppard, P. J., Walter, R., and Nagaoka, T. (2000). The Archaeologyof Head-hunting in Roviana Lagoon, New Georgia. Journal ofthe Polynesian Society 109: 9–37.

Shimada, I. (2000). The late prehistoric coastal states. In Minelli, L. L.(ed.), The Inca World: The Development of Pre-Columbian Peru,1000–1534. University of Oklahoma Press, Norman, OK, pp.49–110.

Spriggs, M. (1997). The Island Melanesians. Blackwell, Oxford, UK.Spriggs, M., and Anderson, A. (1993). Late Colonization of East

Polynesia. Antiquity 67: 200–217.Stevenson, C. M. (1997). Archaeological Investigations on Easter

Island, Maunga Tari: An Upland Agricultural Complex. Bears-ville Mountain Press, Los Osos, CA.

Stock, J., Coil, J., and Kirch, P. V. (2003). Paleohydrology of AridSoutheastern Maui, Hawaiian Islands, and its Implications forPrehistoric Human Settlement. Quaternary Research 59: 12–24.

Sutton, D. G. (1980). A Culture History of the Chatham Islands.Journal of the Polynesian Society 89: 67–93.

Sweeney, M. T. K. (1994). Archaeological Inventory Survey PacificMissile Range Facility, Majors Bay Area, Kaua‘i. Hawai‘i.Report prepared by International Archaeological Research Insti-tute, Inc., Honolulu for Belt Collins Hawaii, Honolulu.

Szabó, K. (2001). Molluscan Evidence for Late Holocene ClimateChange on Motutapu Island, Hauraki Gulf. Journal of thePolynesian Society 110: 79–87.

Takayama, J., and Intoh, M. (1978). Archaeological investigation atChukienu shell midden on Tol, Truk. Nara: Reports of PacificArchaeological Survey 5, Tezukayama University, Japan.

Thomas, F. R. (1995). Excavations at Maunalua Cave, Hawai’i Kai,O’ahu. Hawaiian Archaeology 4: 17–26.

Thomas, F. R., Nunn, P. D., Osborne, T., Kumar, R., Areki, F.,Matararaba, S., Steadman, D., and Hope, G. (2004). RecentArchaeological Findings at Qaranilaca Cave, Vanuabalavu Island,Fiji. Archaeology in Oceania 39: 42–49.

Ulm, S. (2004). Investigations towards a late Holocene archaeology ofaboriginal lifeways on the southern Curtis Coast, Australia. PhDthesis, School of Social Science, University of Queensland,Brisbane, Australia.

Ulm, S., and Hall, J. (1996). Radiocarbon and cultural chronologies insoutheast Queensland prehistory. In Ulm, S., Lilley, I., and Ross,A. (eds.), Australian Archaeology ’95: Proceedings of the 1995Australian Archaeological Association Annual Conference. An-thropology Museum, University of Queensland (Tempus 6), StLucia, pp. 45–62.

Walter, R. (1990). The southern Cook Islands in eastern Polynesianprehistory. Ph.D. dissertation, University of Auckland.

Walter, R. (1996). Settlement Pattern Archaeology in the SouthernCook Islands: A Review. The Journal of the Polynesian Society105: 63–99.

Weisler, M. I. (1995). Henderson Island Prehistory: Colonization andExtinction on a Remote Polynesian Island. Biological Journal ofthe Linnean Society 56: 377–404.

Weisler, M. I. (1996). Taking the mystery out of the Polynesian‘mystery islands’: A case study from Mangareva and the Pitcairngroup. In Davidson, J. M., Irwin, G., Leach, B. F., Pawley, A.,and Brown, D. (eds.), Oceanic Culture History: Essays inHonour of Roger Green. New Zealand Journal of ArchaeologySpecial Publication, pp. 615–629.

Weisler, M. (1999). The Antiquity of Aroid Pit Agriculture andSignificance of Buried A Horizons on Pacific Atolls. Geo-archaeology 14: 621–654.

Weisler, M., and Green, R. C. (2001). Holistic approaches tointeraction studies: A Polynesian example. In Jones, M., andSheppard, P. (eds.), Australasian Connections and NewDirections: Proceedings of the 7th Australasian ArchaeometryConference. Research in Anthropology and Linguistics 5, pp.417–457.

Weiss, H. (2000). Beyond the Younger Dryas: Collapse as adaptation toabrupt climate change in ancient West Asia and the EasternMediterranean. In Bawden, G. and Reycraft, R. M. (eds.),Environmental Disaster and the Archaeology of Human Response.University of New Mexico, Maxwell Museum of Anthropology,Anthropological Papers, No. 7, Albuquerque, pp. 75–95.

Zhang, Q., Zhu, C., Liu, C. L., and Jiang, T. (2005). EnvironmentalChange and its Impacts on Human Settlement in the YangtzeDelta, P.R. China. Catena 60: 267–277.

Hum Ecol