Cultural landscapes on Garua Island, Papua New Guinea.

Cultural landscapes on Garua Island, Papua New Guinea. Biases in Lapita archaeology Most archaeological research on the period characterized by Lapitapottery--dating to c. 3300-1500 cal BP in the Pacific region--hasassumed that villages located on the beach were the primary form ofsettlement (e.g. Kirch 1997: 162, 296) and, by inference, that peoplewere sedentary and had `full-on' agriculture (e.g. Spriggs 1997:88,121). One should be suspicious, however, of the unproblematic linksmade between artefact See artifact. scatters, permanent settlement and agriculture,especially since they are based on dubious analogies to modern villages,which are often products of European colonization. In contrast, Gosden& Pavlides (1994) argued that Lapita sites were visited irregularlyby people practising low-intensity agriculture. It is also possible thatat least some Lapita sites represent locations where pottery and shellvaluables were made and used on special occasions for particular,possibly ritual, purposes. Testing the orthodox package of Lapitavillages, sedentism and agriculture requires considering a broader rangeof land-use models and employing new methodologies to evaluate them. Inthis paper I present a case study to show how adopting the perspectiveof cultural landscapes can lead to a more comprehensive, and thereforemore accurate, view of past human behaviour during the time of Lapitapottery. Cultural landscapes By definition, the term `landscape' takes in all physical andnatural components of the terrestrial environment The Earth's land area, including its manmade and natural surface and sub-surface features, and its interfaces and interactions with the atmosphere and the oceans. . For Pacificarchaeology it should be combined with `seascape' (Gosden &Pavlides 1994) to encompass adequately the settings where humanbehaviour took place. Adding `cultural' to land- and seascapes Seascapes is an RT�� Radio 1 programme broadcast on Fridays at 8.30 pm. and presented by Tom MacSweeney. It is intended to cover all subjects of maritime interest, from leisure to commercial shipping, as well as fishing and the environment. emphasizes the role of the individuals who conceptualized these spacesand actively created and modified them in culturally specific ways (cf.Ashmore & Knapp 1999). The process is interactive: human behaviouris both conditioned by the ideological and physical components ofcultural landscapes and also incorporated within the landscapesthemselves. Since cultural landscapes are the material manifestations ofthe complex interactions between humans and their environment, they arethe ideal focus for archaeological research (Gosden & Head 1994).The terms `social' and `cultural' landscape are both used inthe literature, often to mean roughly the same thing. I prefer thelatter adjective because it is more inclusive and avoids the possibilityof a false dichotomy between the so-called social and utilitarianaspects of behaviour. Although there are numerous archaeological studies of landscape,most have focused either on the mundane aspects of human behaviour, suchas settlement and subsistence (i.e. processual studies), or onlyconsidered the ritual or sacred components (i.e. post-processualstudies) (cf. Knapp and Ashmore in Ashmore & Knapp 1999). Gosden(1989; 1991) introduced the concept of social landscapes to the Pacificregion. Although he considered various kinds of humanbehaviour--conceptions of places, social interaction and mobility--hiswork was nevertheless tied to specific localities identified byarchaeologists by the presence of pottery or other artefact types andcalled `sites', rather than conceiving of the cultural landscape asa whole. As with Gosden's work, many studies assume that one candifferentiate the social or ideological aspects of the landscape andrelate these to a particular set of archaeological components. Theseresearchers have tended to over-emphasize the built environment(especially monuments, standing stones standing stones:see megalithic monuments. , etc.) or rock art. Thesepost-processual studies generally concentrate on places or `sites'within a landscape, rather than on the whole landscape itself (e.g.Ashmore & Knapp 1999). In contrast, studies using`distributional' or `non-site' archaeology (e.g. Rossignol& Wandsnider 1992) have targeted the spatial patterning of artefactsacross large areas, rather than within specific places. These studiestake the individual artefact, rather than the site, as the minimalrecording unit (Holdaway et al. 1998: 1), but they too need to bebroadened to include the spaces between the artefacts and to incorporateother types of data. The ideal archaeology of cultural landscapes should incorporate abroad range of data types and perspectives about past human land use.One solution is the `inherent' view in which ritual and daily lifeare not segregated. Instead, the former is considered to be integral tothe latter (van Dommelen in Ashmore & Knapp 1999). The ideal theoryand methodology, then, will not emphasize particular aspects ofbehaviour over others, but will combine the insights of the ideational i��de��ate?v. i��de��at��ed, i��de��at��ing, i��de��atesv.tr.To form an idea of; imagine or conceive: "Such characters represent a grotesquely blown-up aspect of an ideal man . . . studies of the post-processualists with the total landscape approach ofdistributional archaeology and then go further to take in all elementsof space that have been impacted by and impact on human behaviour. The Garua Island Project A major aim of the archaeological project conducted on GaruaIsland, Papua New Guinea Papua New Guinea(păp`ə, –y (FIGURE 1) has been to monitor changes in humanbehaviour over relatively long periods, rather than concentrate onshort-term events, such as earliest colonization or the short periodwhen Lapita pottery was used. A chronological perspective encompassinglong stretches of time provides a better opportunity for comparing andcontrasting how people have altered their behaviour in relation tochanges in their physical and social environment. The long timeperspective should also make it easier to tease out the particular fromthe general (e. g. whether the introduction of Lapita pottery is aunique event or a continuation of the maintenance of long-distanceties). Research on Garua Island has tried to understand how plantutilization, strategies of forest management, mobility patterns, rawmaterial procurement, artefact manufacture and use, social interactionincluding exchange, and cultural constructions of landscape alteredthroughout the Holocene period. [FIGURE 1 OMITTED] To date, interactions between people and their landscapes on GaruaIsland have been monitored through 1 environmental reconstruction using geology and geomorphology geomorphology,study of the origin and evolution of the earth's landforms, both on the continents and within the ocean basins. It is concerned with the internal geologic processes of the earth's crust, such as tectonic activity and volcanism that constructs new combined with dating (Torrence et al. 2000; Torrence & Stevenson2000); 2 analysis of sediments to detect human causes and effectsconcerning pedogenic, geomorphic ge��o��mor��phic?adj.Of or resembling the earth or its shape or surface configuration. and volcanic processes (Boyd &Torrence 1996); 3 ecofacts such as starch grains (Therin et al. 1999) andphytoliths (Boyd et al. 1998; Lentfer et al. 2001; Parr et al. 2001); 4 artefacts, including analyses of obsidian characterization(Torrence & Summerhayes 1997), technology (Torrence et al. 2000),use-wear and residues (Barton et al. 1998; Kealhofer et al. 1999) anddistributional patterning (Torrence et al. 2000); and 5 deliberate human markings of places such as cup marks and rockart All these aspects should be woven together to create anunderstanding of cultural landscape history. In this paper I illustratejust one set of methods--the distributional study of obsidianartefacts--in order to demonstrate the value of looking beyond the site. Buried landscapes `Distributional', `off-site', `non-site'archaeologies, all of which concentrate on the interpretation ofartefact distributions across ground surfaces (e.g. Ebert 1992;Rossignol & Wandsnider 1992; Holdaway et al. 1998), would seem to beideal methodologies for the study of cultural landscapes because theyfocus on the broadest spatial scale. The basic assumption of theseapproaches is that patterns of artefact discard are largely a product ofthe way people construct their landscapes, move around, subsist sub��sist?v. sub��sist��ed, sub��sist��ing, sub��sistsv.intr.1. a. To exist; be.b. To remain or continue in existence.2. andinteract with each other and their environment. A very real problem forthe Pacific region is dense vegetation that makes surface surveyvirtually impossible except on beaches and reefs. A more importantlimitation is that most modern surfaces have been exposed for some timeand therefore have had a very long and complex history (e.g. Wandsniderin Rossignol & Wandsnider 1992; Holdaway et al. 1998). Furthermore,many surfaces are relatively unstable and, when combined with human landuse, are subject to erosion and redeposition Noun 1. redeposition - deposition from one deposit to anotherdeposition, deposit - the natural process of laying down a deposit of something (e.g. Gosden & Webb1994; Kirch & Hunt 1996). Consequently, the complex palimpsest palimpsest(păl`ĭmpsĕst'): see manuscript. ofartefacts on most ground surfaces means the interpretation of theirrelation to human behaviour, especially in terms of short-term events,is problematic and may be quite limited (Wandsnider 1992). Garua Island offers a unique possibility for overcoming some of thetaphonomic nightmares that beset landscape archaeology Landscape archaeology is a body of method and theory for the study of the material traces of past peoples within the context of their interactions in the wider (typically regional) social and natural environment they inhabited. . As shown inFIGURE 2, ancient ground surfaces, each of which had a relatively shortlife, were sealed by a layer of airfall tephra teph��ra?n.Solid matter that is ejected into the air by an erupting volcano.[Greek tephr (volcanic ash See under Ashes.See also: Ash ) (Torrenceet al. 2000). The distinctive and dateable tephras effectively isolateone time slice A short interval of time allotted to each user or program in a multitasking or timesharing system. Time slices are typically in milliseconds. (operating system) time slice from the next. The artefacts preserved within the buriedsoils allow us to reconstruct landscapes that were created and modifiedover a moderate length of time. This is the good news for distributionalapproaches. The bad news is that plotting the position of individualartefacts over large areas, as in the classic surface-distributionstudies (e.g. Ebert 1992; Holdaway et al. 1998), is not feasible. Whenone has to dig down to undermine and cause to fall by digging; as, to dig down a wall.See also: Dig almost two metres to reach soils dated to c. 5000years ago, time, energy and the tolerance of land owners grossly limithow much area can be exposed. Sampling is therefore a necessity. [FIGURE 2 OMITTED] A one-metre square area of land surface obtained by excavationusing natural stratigraphic stra��tig��ra��phy?n.The study of rock strata, especially the distribution, deposition, and age of sedimentary rocks.strat layers comprises the basic sample.Geomorphological ge��o��mor��phol��o��gy?n.The study of the evolution and configuration of landforms.geo��mor analysis determined if the sediments were in situ In place. When something is "in situ," it is in its original location. (Boyd& Torrence 1996). In total 177 landscape samples dated to threeperiods were recovered from 69 test pits. Use-wear samples werecollected from each stratigraphic unit. Column samples for plantmicrofossils were taken from a large proportion of the pits. This paperpresents an interpretation of changes in cultural landscapes based onthe relative density of obsidian artefacts associated with the buriedground surfaces. Tephra stratigraphy stratigraphy,branch of geology specifically concerned with the arrangement of layered rocks (see stratification). Stratigraphy is based on the law of superposition, which states that in a normal sequence of rock layers the youngest is on top and the oldest on the During the past 5900 years, fails of airfall tephra from threevolcanic events buried and sealed a series of landscapes on Garua Island(Torrence et al. 2000). Some of the tephra layers are quite thin and/ordiscontinuous discontinuous/dis��con��tin��u��ous/ (dis?kon-tin��u-us)1. interrupted; intermittent; marked by breaks.2. discrete; separate.3. lacking logical order or coherence. and so, for the purpose of analysis, I have concentratedon three well-preserved buried soil horizons. Radiocarbon and obsidianhydration hydration/hy��dra��tion/ (hi-dra��shun) the absorption of or combination with water. hy��dra��tionn.1. The addition of water to a chemical molecule without hydrolysis.2. dates have substantiated the stratigraphic interpretations(cf. Torrence et al. 2000; Torrence & Stevenson 2000). Therelationship between the stratigraphy and chronological periods is shownin FIGURE 3. [FIGURE 3 OMITTED] Period I (5900-3600 cal BP) is placed between the W-K2 and W-K1tephras, but since W-K1 was rarely preserved in situ, a 20-cm thicksample of the soil under W-K2 was used as a best approximation based onradiocarbon dates from preliminary excavations. Period II (3400-1000 cal BP) is situated between the W-K2 and Dktephras and includes the time of Lapita pottery. Period III (1000 cal BP-present) is above the Dk tephra. Deposition on the three ground surfaces formed on the 3 tephrastook place over slightly different lengths of time, with Period IIIshorter than the others. In terms of most surface archaeology, however,the length of time incorporated in each buried soil is very short (cf.Sullivan 1998). Sampling The spatial distribution of the sample test pits is presented inFIGURE 4. In order to ensure that the buried Garua landscapes werecomprehensively and efficiently sampled given the resources available, Iemployed judgmental judg��men��tal?adj.1. Of, relating to, or dependent on judgment: a judgmental error.2. Inclined to make judgments, especially moral or personal ones: sampling within a stratified stratified/strat��i��fied/ (strat��i-fid) formed or arranged in layers. strat��i��fiedadj.Arranged in the form of layers or strata. design with geology,stream catchments, physiography and topography defining the strata. Thevalue of this approach is that it directed research to zones which hadpreviously been ignored, especially the inland portion of the island.The first sampling frame is composed of the two basic geological unitsof the island. These correspond with the Hamilton and Baki peaks whichare separated by a major stream catchment (FIGURE 4). The raised coralexposed within the modern stream bed indicates that the Hamilton sidewas once a separate island. Obsidian outcrops on each side of the islandwere formed in different types of rhyolite rhyolite,fine-grained light-colored acidic volcanic rock. Rhyolite is chemically the equivalent of granite, and is thus composed primarily of quartz and orthoclase feldspar with subordinate amounts of plagioclase feldspar, biotite mica, amphiboles, and pyroxenes. and have distinctivelydifferent physical appearances, flaking properties and chemicalcompositions (e.g. Torrence et al. 1992; Bird et al. 1997). [FIGURE 4 OMITTED] Secondly, five stream catchments were sampled. Thirdly, thephysiographic phys��i��og��ra��phy?n.See physical geography.physi��ogra��pher n. zones were sampled in roughly equal proportions to theiroccurrence, with the exception of the steep upland which wasunder-represented because buried soils were poorly preserved (TABLE 1).Fourthly, topographic features identified during the course of thefieldwork were sampled in the relative frequencies listed in TABLE 2.Steep slopes and valley bottoms were sampled at a lower frequency thantheir occurrence because erosion had removed many buried horizons. Three transects were selected to simplify transport of people andequipment. As shown in FIGURE 4, two north-south transects sample theinterior regions: one runs across the western side of Mt Baki andanother was placed to the east of Mt Hamilton. A third transect runseast-west from the coast to the top of Mt Hamilton and provides a directcomparison of coast and interior regions. The coastal plain on the southwestern part of the island was intensively sampled because elsewherethis zone largely consists of recent alluviation. Obsidian artefact distributions Nearly all the test pits produced large quantities of obsidianchipped stone In archaeology, chipped stone refers to a method of manufacturing stone tools through lithic reduction, wherein lithic flakes are struck off a mass of tool stone with a percussor. artefacts. Also surprising was the relatively widespreaddistribution of Lapita pottery and the large numbers of potsherds inunusual settings, particularly some inland hill and ridge tops (FIGURE5; cf. Torrence & Stevenson 2000). There are virtually no emptypatches on these landscapes. Only one test pit (B1), located in a valleywith high runoff, had low counts of artefacts for all periods. Oneastute volunteer asked how the past barefoot residents avoided seriouslycutting their feet on the abundant sharp obsidian flakes that seem tocarpet the landscapes. [FIGURE 5 OMITTED] To get a preliminary impression of the distribution of obsidianartefacts, raw counts were transformed to numbers per 0.1 cubic metre Noun 1. cubic metre - a metric unit of volume or capacity equal to 1000 literscubic meter, kiloliter, kilolitremetric capacity unit - a capacity unit defined in metric terms .These were then divided into low (0-5), medium (6-60) and high (>60)classes. TABLE 3, which summarizes the occurrence of density classes foreach period, highlights some very intriguing patterns. (Since not allburied landscapes are preserved in every test pit, totals vary.)Artefacts are most abundant in Period I with high counts for 40 pits,comprising 65% of the total, and only 4 low examples. As shown in FIGURE6a, test pits containing high densities of artefacts are continuouslydistributed across the landscape with no special preference forphysiographic type. There are no empty spaces, no significant clustersand little preference regarding situation in terms of inland or coastallocation. [FIGURE 6 OMITTED] It is worth noting that during Period I the sea level reached thebase of the coastal cliffs and most of the current coastal plain did notexist. The two coastal pits with low counts, FCY FCY Foreign Currency II and FAS II (FIGURE4), are located in areas where active fans were operating and thematerial may therefore have been redeposited. In any case, these wereprobably not favourable settings for human activity. If these cases areeliminated, there are only 2 low-density locations: the very top of MtHamilton and a very damp valley bottom. In summary, during Period Ilarge quantities of obsidian were utilized over the whole island and theentire landscape was treated equally in terms of artefact discard. Byimplication, the island was conceived of as uniform for at least someactivities. Period II (FIGURE 6b) represents a significant shift in behaviourbecause the relative size of the low density class is much larger thanin the previous period and cases are more evenly spread among theclasses (cf. chi-square results in TABLE 4). The smaller number ofpreserved contexts in Period II reflects higher rates of erosion,possibly because of the instability of the W-K2 tephra, but this isunlikely to have affected the relative distribution among the densitycategories. Unlike the previous period, there is a much lower relativenumber of samples with large quantities of obsidian and, significantly,a much higher percentage of places have low rates of discard. These datarepresent a trend in the discard of obsidian artefacts from an evenspread to somewhat discrete clusters surrounded by areas with lowdensities of material. The majority of high density samples are located in the inlandzones on hills or ridge tops, although most of these have a good view ofthe sea. This pattern was not expected, because the settlement patternof sites with Lapita pottery, which date to this period, has beenstrongly linked to beach settings in other regions (e.g. Kirch 1997:162). The three high-density test pits located on the coastal plain (D5,D6, D7) are quite close together and may represent samples of a singlehigh-density scatter. Two other test pits near the coast contained veryfew artefacts (FCY II, FAS II; cf. FIGURE 4). Furthermore, the uplandplain around the base of the volcanic cone a hill, conical in form, built up of cinders, tufa, or lava, during volcanic eruptions.See also: Volcanic that makes up Mt Hamilton hasvery little material, unlike the previous period. To sum up, in PeriodII one can identify a more clustered distribution than Period I,movement onto the new coastal plain, a focus on ridges and hilltops anda tendency to avoid the high interior regions of the island. Period III continues the trend toward a clustered distribution ofmaterial. There is a smaller percentage of cases in the high-densitycategory and a larger proportion in the low group, although the patternis not significantly different from Period II as measured by thechi-square statistic (TABLE 4; cf. FIGURE 6c). What is most striking,however, is that the high-density samples are all in upland regions andmore than half do not have a direct view of the sea, whereas the low ormedium densities are coastal. This represents a significant refocusingof some activities to an area with little use in the previous period. In summary, in the three periods examined, the spatial patterns ofobsidian densities within the test pits vary markedly. In Period I thereis a homogenous homogenous - homogeneous spread of obsidian across the landscape with all areascontaining large amounts of obsidian. In contrast, Period II ischaracterized by a slightly clustered pattern and the majority of placeswith the highest rates of discard avoid the interior and, although notsituated on the beach, appear to be oriented toward the coast. PeriodIII continues the trend toward a clustered pattern with increasinglyempty spaces in between, suggesting that through time activitiesinvolving stone artefacts were increasingly restricted in space.Finally, in Period III the bulk of obsidian discard moved from coastalhills and ridge to the interior portions of the island. Change in cultural landscapes Significant differences in how people created their culturallandscapes on Garua Island are clearly reflected by the distribution ofobsidian among the test pits as well as their spatial patterning acrossthe island. Preliminary hypotheses for these shifts emphasizesubsistence and mobility patterns and the nature of social conflict.Variation in the clustered nature of artefact discard can be correlatedwith temporal patterning in raw material procurement and lithictechnology In archeology, Lithic Technology refers to a broad array of techniques and styles to produce usable tools from various types of stone. The earliest stone tools were recovered from modern Ethiopia and were dated to between two-million and three-million years old. which together indicate a reduction in mobility, probablycaused by an increase in the intensity of land management (Torrence etal. 2000; cf. Torrence in Galipaud 1992). From at least 10,000cal. BP,gearing-up of multi-purpose obsidian tools took place at very specificraw material sources and was backed up by continuous maintenance asgroups moved over relatively large areas practicing a low intensity formof land management. Through time this pattern of behaviour graduallybecame more expedient. In Periods II and III stone obtained throughlocal exchange systems was stockpiled at places which were occupied fora longer time because of a shift to more intense forms of cultivation.Use-wear residue studies of sites on the mainland (Fullagar in Galipaud1992; Kealhofer et al. 1999) and Pavlides' (1999) study of theinterior of West New Britain Coordinates: West New Britain is a province of Papua New Guinea on the islands of New Britain. The provincial capital is Kimbe. also support this general picture. The test-pit data indicate a continuous change from a widespreaddistribution of activities that could be associated with multiple,short-term activities, to one that was more focused on a few particularplaces, which might be reconstructed as longer-term occupations. It isquite significant that, although the major change took place betweenPeriods I and II, the trend towards localized obsidian working continuedinto Period III. One might infer from these results that the pattern ofland use during the time of Lapita pottery was less sedentary than inthe most recent 1000 years, thereby providing support for Gosden &Pavlides' (1994) model. It is interesting to compare the preliminary results from GaruaIsland with those from Gosden's (1989; 1991; Gosden & Pavlides1994) landscape studies for the Arawe Islands region on the south coastof New Britain New Britain, city, United StatesNew Britain,industrial city (1990 pop. 75,491), Hartford co., central Conn.; settled c.1686, inc. 1871. The tin shops and brassworks in the city were established in the 18th cent. . He found that the Lapita sites represented relativelylarge concentrations of material located on the coast, whereas in thelater period smaller hamlets were dispersed within the inland. On GaruaIsland clustering of obsidian artefacts is less tight during the time ofLapita pottery, Period II, than in the subsequent period. If, however,one compares the overall spatial distribution of individual test pitswith high densities, they are more dispersed in Period III than inPeriod II, but the difference is not a major one (FIGURES 6b, 6c).However, the shift from coastal to inland occupation that Gosden hasidentified is paralleled on Garua Island. Period II test pits on Garua with high densities of obsidian arefound on coastal hill and ridge tops (cf. Torrence & Stevenson 2000)rather than on the beach as in the Arawe Islands (Gosden 1989: 52;Gosden & Webb 1994) and elsewhere (Kirch 1997: 162). The lack ofLapita pottery' in original beach contexts on Garua may, however,be partly a product of erosion following recent uplift, but the spreadof chipped stone and pottery throughout much of the island at this timeis a pattern that is practically unknown outside this region. Morerecent work targeting cultural landscapes on the adjacent mainland hasalso recovered Lapita pottery in inland settings and on isolatedhilltops (cf. Torrence & Stevenson 2000). The change between Periods II and III in the location of thedensest scatters of obsidian artefacts (coastal to inland) is difficultto explain solely in terms of subsistence, since the distance betweendense scatters and different resources is not great enough to haveaffected travel costs significantly. At the beginning of both periodsGarua was reoccupied following a major volcanic eruption, but each timepeople chose to focus their activities in different physiographicregions. During the time of Lapita pottery, artefact discard graduallymoved inland (Torrence & Stevenson 2000), but the foothills of MtHamilton, which were heavily used in Period III, were avoided in PeriodII. In Period II the high-density scatters were usually placed on smallhill- and ridge-tops which have well defined steep edges, whereas inPeriod III they were located in the interior of the island, but notnecessarily in such obvious defensible settings. The locational, asopposed to the structural, changes between Periods II and III canperhaps be explained in terms of changes in the nature of socialinteraction and/or ideological conceptions of appropriate places forcertain classes of behaviour. For example, the marked decrease inimported obsidian in Period III implies a reorientation Noun 1. reorientation - a fresh orientation; a changed set of attitudes and beliefsorientation - an integrated set of attitudes and beliefs2. reorientation - the act of changing the direction in which something is oriented of exchangerelationships (Torrence & Summerhayes 1997), which in turn may belinked to changes in the nature of social conflict or warfare. Although the long-term trend on Garua Island is for a shift towardfewer places with high densities of obsidian, limiting the term`site' to these situations is inappropriate since material was alsodiscarded outside these contexts throughout the past 5000 years atleast. Although Lapita pottery is always found in association withrelatively abundant quantities of obsidian artefacts, the number ofsherds at some inland locations is very small (Torrence & Stevenson2000), raising the question of what constitutes a Lapita `site'(White in Galipaud 1992). In addition, during Period II there are largequantities of obsidian at places where pottery is absent, suggestingpottery was only discarded in particular contexts, whereas the disposalof obsidian may have involved other activities and carried othermeanings. It is clear that throughout the past people moved across andused the entire island, but the cultural landscapes that they createddemonstrate that they did this in different ways. Broader implications This first-stage analysis of the distributional data collected onGarua Island demonstrates that focusing on cultural landscapes has thepotential to alter our understanding of the prehistory prehistory,period of human evolution before writing was invented and records kept. The term was coined by Daniel Wilson in 1851. It is followed by protohistory, the period for which we have some records but must still rely largely on archaeological evidence to of Near Oceania Near Oceania is the part of Oceania comprising Australia, New Guinea and the Solomon Islands. The other part of Oceania is Remote Oceania. Most linguists and scientists consider this a natural division of Oceania. and perhaps further afield. Firstly, the data conclusively demonstratethat, during the past c. 6000 years, the discard of chipped stoneartefacts was not limited to specific, circumscribed places which matchthe definition of a `site' as a well-defined and bounded entity.Significant quantities of material were discarded over much of the landsurface and differences in the relative occurrence and spatialpatterning of chipped stone artefacts indicate changes in how peopleused and created their cultural landscapes. Secondly, the long-term perspective of the Garua research isimportant because a better understanding of the period with Lapitapottery is gained by comparing it with processes operating before andafterwards. The patterns I have identified suggest a gradual reductionin mobility and, by inference, variations in land-management practices.The major change between Periods I and II suggests a different form ofsubsistence pattern, perhaps linked to the introduction of new plantsand animals, as proposed by a number of scholars (e.g. Spriggs 1997). Itis important to note, however, that neither the Lapita system of landuse nor social system were stable, and so making direct links betweencurrent Pacific societies and their ancestors is more problematic thanmany scholars admit. Thirdly, the Garua case study illustrates that important data abouthuman land use are lost when only a few select points are targeted foranalysis. If one had taken the traditional approach to Pacificarchaeology and only excavated small areas at a few coastal localities,then a very biased picture of changes in how people viewed and usedtheir environments would have been obtained. The distributional studydemonstrates that human activity was not always concentrated on thebeach or even within the coastal zone, as has so often been assumed,particularly for the Lapita period. The Garua data also show that theinland zone was used extensively during the time of Lapita pottery, thatit was visited and used throughout pre-history, and that it was thefocus for activity particularly in Period I, when the existing coastalplain did not exist, and again in Period III. The large quantity ofobsidian discarded in the interior of the island also indicates thatterrestrial resources played an important role in all periods.Archaeological research in Near Oceania therefore needs to alter themethodology used to find and recover archaeological contexts and torecast interpretations of behaviour to include activities taking placeaway from the coast. Fourthly, expanding the temporal and spatial framework of theresearch removes the assumption that in every case excavated materialfound outside rock shelters is derived from `villages' where allsignificant activities took place. Finally, shifting the focus from sites to cultural landscapesshould encourage archaeologists to look beyond subsistence andsettlement as the most important causal factors and to integrate socialand ceremonial constructions of space, as illustrated by the discussionof changes between Periods II and III. As I emphasized previously,`cultural' should be defined to encompass all aspects of behaviourwith no strict separation between the mundane and ideological. The analysis of the Garua Island test-pit data presented here isbased on a fraction of what will be learned by studying the material inmore detail. For example, a better understanding of the activities thatgenerated the scatters of obsidian artefacts will be gained throughtechnological and use-wear/residue studies still in progress. It will beinteresting to compare and contrast the differences in activitiesresponsible for the discard of material within the high-, medium- andlow-density scatters and between inland versus coastal settings. Furtherdata are required to test the hypothesis that through time there was agradual reduction in mobility combined with an increase in the intensityof land management or that changes in social conflict took place betweenPeriods II and III. Methods for integrating the rock art are alsorequired. Garua Island provided a very important opportunity to capitalize on Cap´i`tal`ize on`v. t. 1. To turn (an opportunity) to one's advantage; to take advantage of (a situation); to profit from; as, to capitalize onan opponent's mistakes s>. the advantages of studying cultural landscapes. Certainly, the presenceof easily recognized, well-defined and short-lived buried groundsurfaces is a major benefit that cannot be replicated in other areas ofthe Pacific where high rates of erosion have remodelled prehistoriclandscapes. An added advantage is the proximity to obsidian sources. Theabundance of obsidian artefacts on Garua provides an excellent `litmus litmus,organic dye usually used in the laboratory as an indicator of acidity or alkalinity (see acids and bases). Naturally pink in color, it turns blue in alkali solutions and red in acids. paper' for detecting the presence of human activities that mightnot be so visible in situations where raw material was more carefullyconserved. The archaeology of cultural landscapes using a distributionalstudy of material preserved on buried landscapes has already providedsome unexpected and important results and has opened up exciting andimportant new areas of research. Although Garua Island has specialproperties, the advantages of studying a wide range of landforms andshifting the emphasis from sites (assumed to be foci of behaviour) toactivities wherever they occur could be shared by many otherarchaeologists.TABLE 1. Distribution of test pits by physiographicregion.physiographic region number %coastal plain 6 9coastal escarpment 11 16coastal cliffs 4 6upland, plain 42 61upland, steep 6 9total 69 101TABLE 2. Distribution of test pits by topography.topography number %hilltop 4 6hillslope 9 13ridge top 13 19ridge spur 9 13ridge slope 14 20saddle 6 9valley head/bottom 3 4hollow 2 3level ground 9 13total 69 100TABLE 3. Relative abundance of obsidian. number of test pits Period I Period II Period IIIobsidianabundance no. % no. % no. %high 40 65 20 39 14 22medium 18 29 19 37 29 45low 4 6 12 24 21 33total 62 51 64 Acknowledgements. The research was funded by the AustralianResearch Council, Australia and Pacific Foundation, Australian Museum,and Earthwatch Institute. Additional support came from the NationalMuseum and Art Gallery, National Research Institute, West New BritainProvincial Government and Cultural Centre, Kimbe Bay Shipping Agenciesand Walindi Plantation. Thanks are extended to the international team ofvolunteers. I am especially grateful to Jo Bola bo��la? also bo��lasn.A rope with weights attached, used especially in South America to catch cattle or game by entangling their legs.[From American Spanish bolas, pl. , Nick Lyons, Les Hartwigand their staff on Garua Plantation and the people of West New Britainfor their assistance, hospitality and friendship. References ASHMORE, W. & A. KNAPP (ed.). 1999. Archaeologies of landscape.London: Blackwell. BARTON, H,, R. FULLAGAR & R. TORRENCE. 1998. Clues to stonetool function re-examined: comparing starch grain frequencies on usedand unused obsidian artefacts, Journal of Archaeological Science 25:1231-8. BIRD, J.R., R. TORRENCE, G. SUMMERHAYES & G. BAILEY. 1997. NewBritain obsidian sources, Archaeology in Oceania 32: 54-60. BOYD, W.E., C.J. LENTFER & R. TORRENCE. 1998. Phytolith phy��to��lith?n.A minute particle formed of mineral matter by a living plant and fossilized in rock. analysis for a wet tropics tropics,also called tropical zone or torrid zone, all the land and water of the earth situated between the Tropic of Cancer at lat. 23 1-2°N and the Tropic of Capricorn at lat. 23 1-2°S. environment: methodological issues andimplications for the archaeology of Garua Island, West New Britain,Papua New Guinea, Palynology pal��y��nol��o��gy?n.The scientific study of spores and pollen.[Greek palunein, to sprinkle + -logy. 22: 213-28. BOYD, W.E. & R. TORRENCE. 1996. Periodic erosion and humanland-use on Garua Island, PNG (Portable Network Graphics) A bitmapped graphics file format endorsed by the World Wide Web Consortium. It is expected to eventually replace the GIF format, because there are lingering legal problems with GIFs. : a progress report, in S. Ulm, I. Lilley& A. Ross (ed.), Australian Archaeology '95, Tempus 6: 265-74.Brisbane: University of Queensland The University of Queensland (UQ) is the longest-established university in the state of Queensland, Australia, a member of Australia's Group of Eight, and the Sandstone Universities. It is also a founding member of the international Universitas 21 organisation. . EBERT, J. 1992. Distributional archaeology. Albuquerque (NM):University of New Mexico Press The University of New Mexico Press, founded in 1929, is a university press that is part of the University of New Mexico. External linkUniversity of New Mexico Press . GALIPAUD, J.-C. (ed.). 1992. Poterie Lapita et Peuplement. Noumea:ORSTOM ORSTOM Office de la Recherche Scientifique et Technique d'Outre-Mer (French). GOSDEN, C. 1989. Prehistoric social landscapes in the AraweIslands, West New Britain Province, Papua New Guinea, Archaeology inOceania 24: 45-58. 1991 Toward s an understanding of the regional archaeologicalrecord from the Arawe Islands, West New Britain Province, Papua NewGuinea, in J. Allen & C. Gosden (ed.), Report of the Lapita HomelandProject: 205-16. Canberra: Department of Prehistory, Research School ofPacific Studies, Australian National University. Occasional Papers inPrehistory 20. GOSDEN, C. & L. HEAD. 1994. Landscape--a usefully ambiguousconcept, Archaeology in Oceania 29: 113-16. GOSDEN. C. & C. PAVLIDES. 1994. Are islands insular? Landscapevs seascape in the case of the Arawe Islands, Papua New Guinea,Archaeology in Oceania 29: 162-71. GOSDEN, C. & J. WEBB. 1994. The creation of a Papua New Guineanlandscape: archaeological and geomorphological evidence, Journal ofField Archaeology 21: 29-51. HOLDAWAY, S., D. WITTER, P. FANNING, R. MUSGRAVE, G. COCHRANE, T.DOELMAN, S. GREENWOOD, D. PIGDON & J. REEVES. 1998. New approachesto open site archaeology in Sturt National Park Sturt is a national park in New South Wales (Australia), 1059 km northwest of Sydney. See alsoProtected areas of New South Wales (Australia) , New South Wales New South Wales,state (1991 pop. 5,164,549), 309,443 sq mi (801,457 sq km), SE Australia. It is bounded on the E by the Pacific Ocean. Sydney is the capital. The other principal urban centers are Newcastle, Wagga Wagga, Lismore, Wollongong, and Broken Hill. ,Australia, Archaeology in Oceania 33:1-19 KEALHOFER, L., R. TORRENCE & R. FULLAGAR. 1999. Integratingpbytoliths within use-wear studies of stone tools, Journal ofArchaeological Science 26: 527-46. KIRCH, P. 1997. The Lapita peoples: ancestors of the Oceanic world.London: Blackwell. KIRCH, P. & T. HUNT (ed.). 1996. Historical ecology in thePacific Islands: prehistoric landscape and environmental change. NewHaven (CT): Yale University Press. LENTFER, C., B. BOYD & R. TORRENCE. 2001. Phytolith researchrelating to the archaeology of West New Britain, Papua New Guinea, inJ.M. Meunier & F. Calin (ed.), Phytoliths: applications in earthsciences and human history: 213-24. Tokyo: A.A. Balkema. PARR, J., C. LENTFER & W. BOYD. 2001. Spatial analysis offossil phytolith assemblages at an archaeological site in West NewBritain, Papua New Guinea, in G. Clark, A. Anderson & T. Vunidilo(ed.), The archaeology of Lopita dispersal in Oceania: 125-34. Canberra:Pandanus Books. PAVLIDES, C. 1999. The story of Imlo: the organization of flakedstone technologies from the lowland tropical rainforest of West NewBritain, Papua New Guinea. Unpublished Ph.D thesis, School of Historyand Archaeology, La Trobe University 1. u/r = unranked2.AsiaWeek is now discontinued. Student lifeDuring the 1970s and 1980s, La Trobe, along with Monash, was considered to have the most politically active student body of any university in Australia. . ROSSIGNOL, J. & L. WANDSNIDER (ed.). 1992. Space, time, andarchaeological landscapes. New York New York, state, United StatesNew York,Middle Atlantic state of the United States. It is bordered by Vermont, Massachusetts, Connecticut, and the Atlantic Ocean (E), New Jersey and Pennsylvania (S), Lakes Erie and Ontario and the Canadian province of (NY): Plenum. SPRIGGS, M. 1997. The Island Melanesians. Oxford: Blackwell. SULLIVAN, III, A.P. (ed.) 1998. Surface archaeology. Albuquerque(NM): University of New Mexico Press. THERIN, M., R. FULLAGAR & R. TORRENCE. 1999. Starch insediments: a new approach to the study of subsistence and land use inPapua New Guinea, in C. Gosden & J. Hather (ed.), Food inprehistory: 438-62. London: Routledge. TORRENCE, R., C. PAVLIDES, P. JACKSON & J. WEBB. 2000. Volcanicdisasters and cultural discontinuities in the Holocene of West NewBritain, Papua New Guinea, in B. McGuire, D. Griffiths & I. Stewart(ed.), The archaeology of geological catastrophes: 225-44. London:Geological Society of London The Geological Society of London is a learned society based in the United Kingdom with the aim of "investigating the mineral structure of the Earth". It is the oldest national geological society in the world and the largest in Europe with over 9000 Fellows entitled to the . Special Publication 171. TORRENCE, R., J. SPECHT, R. FULLAGAR & R. BIRD. 1992. FromPleistocene to present: obsidian sources in West New Britain, Papua NewGuinea, Records of the Australian Museum, Supplement 15: 83-98. TORRENCE, R. & C. STEVENSON. 2000. Beyond the beach: changingLapita landscapes on Garua Island, in T. Murray & A. Anderson (ed.),Australian archaeologist: papers in honour of Jim Allen: 324-45.Canberra: Coombs Academic Publishing. TORRENCE, R. & G. SUMMERHAYES. 1997. Sociality and the shortdistance trader: intra-regional obsidian exchange in the Willaumezregion, Papua New Guinea, Archaeology in Oceania 32: 74-84. ROBIN TORRENCE, Division of Anthropology, Australian Museum, 6College Street, Sydney College Street in Sydney, New South Wales, Australia is a major street in the centre of the Central Business District. It runs from Queens Square near St James station to Whitlam Square at Liverpool St. NSW NSWNew South WalesNoun 1. NSW - the agency that provides units to conduct unconventional and counter-guerilla warfareNaval Special Warfare 2010, Australia. robint@austmus.gov.au Received 1 February 2001, accepted 8 October 2001, revised 8 March2002