sign in sign up
<<
Home , Arorae, Choiseul Island, Enderbury Island, Kanton Island, Manra, Nikumaroro

Vol. 147, No. 2 · Research article

Only a pawn in their games? DIE ERDE environmental (?) migration in Journal of the Geographical Society – past, present and future of Berlin

Eberhard Weber1

1 School of Geography, Earth Science and Environment, Laucala Bay Road, Suva, , [email protected]

Manuscript submitted: 14 August 2015 / Accepted for publication: 11 April 2016 / Published online: 30 June 2016

Abstract The Pacific Countries and Territories (PICT) are exposed to the impacts of climate change. In extreme cases entire states may disappear. Kiribati is one of these countries. Within its own territory there are no ­places to where people could be safely resettled when their home islands become unsuitable for human habitation. Large-scale resettlement is nothing new to the people of Kiribati. In colonial times people from various islands were resettled. The Phoenix Island Settlement Scheme (PISS) is one of these efforts to allegedly bring people to safety. Making use of primary sources that have become available only recently the paper raises the ques- tion if there is anything to learn from PISS for present times, or if PISS has historical value only, as the ­’s last colonial expansion scheme. The paper asks about conflicting intentions of colonial authorities and assesses if and possibly why strategic political considerations resulted in a situation where humanitarian motivations retreated into the background leading to a sub-optimal preparation of the scheme, which then fi- nally led to its failure. The paper comes to the conclusion that behind reportedly noble purposes there is a layer of colonial interests which lets settlers appear as objects in a larger colonial game.

Zusammenfassung

- Die Pazifischen Inselstaaten und -territorien sind den Auswirkungen des globalen Klimawandels ausgesetzt. In extremen Fällen kann dies dazu führen, dass ganze Staaten verschwinden. Kiribati ist eines dieser Län- der. Innerhalb des eigenen Staatsgebietes gibt es keine Orte, wohin Menschen in Sicherheit gebracht werden umgesiedelt.könnten, sollten Das ihre Phoenix Heimatinseln Island Settlement unbewohnbar Scheme werden. (PISS) Große Umsiedlungsmaßnahmen sind nichts Außer gewöhnliches für die Menschen in Kiribati. In der Kolonialzeit wurden Menschen von verschiedenen Inseln ist eine dieser Bemühungen, Menschen vermeintlich in Sicherheit zu bringen. Mit Hilfe von Primärquellen, die erst kürzlich der Öffentlichkeit zugänglich gemacht wurden, wirft der Betrag die Frage auf, ob man von PISS für die Gegenwart und Zukunft etwas lernen kann, oder ob die Auseinandersetzung mit dieser letzten kolonialen Expansion des britischen Weltreiches lediglich historische Neugierde befriedigt. Der Beitrag geht sich widersprechenden Absichten der Kolonialbehörden nach und versucht herauszufinden, weshalb möglicherweise strategisch-politische Erwägungen dazu führen konnten, dass humanitäre Motive in den Hintergrund traten und zu einer unzureichenden Vorbereitung des Vorhabens führten, was am Ende das Scheitern des Projektes bewirkte. Der Beitrag kommt zu dem Schluss, dass sich hinter vorgeblich noblen Absichten eine Reihe kolonialer Interessen verbarg, die die Kolonisten als KeywordsObjekte in einer größeren kolonialen Partie erscheinen lassen.

Weber, Eberhard Kiribati,2016: Only colonial a pawn in resettlement,their games? environmental environmental (?) migration and climate in Kiribati change – past, present and future. – DIE ERDE 147 (2): 153-164

DOI: 10.12854/erde-147-11

DIE ERDE · Vol. 147 · 2/2016 153 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

1. Introduction -

- andSuch mobility, historical but cases still seem they toraise connect a number well withof issues con temporary discourses about environmental change Mainstream science today is confident that the ac - cumulation of greenhouse gases in the atmosphere when looking closer. PICTs are consideredBarnett to be and at causes dangerous climate change that put – among Adgerhigh risk to various impacts of climate change. At other impacts – people’s lives and livelihoods at risk.- olls in the Pacific are especially exposed ( Developing countries are Barnett most exposed and Adger to 2003; these 2003). In worst case scenarios atollsMcAdam disappear (2014, McDowell­impacts, butet al. least 2016; prepared Weber 2014a). and with little capaci within the next few decades and countries cease ties to face the challenges ( to exist. Under such assumptions - - 2015) looks at planned relocation and resettlement thatreferring e.g. have to colonial low population cases and densitiesintentions: or to are get unin peo- A small but important part of climate change dis ple out of ‘danger zones’ and take them to ‘safe’ areas courses is about mobility: bringing people to safety,- Lieber or supporting them in their own efforts to reach habited, leaving ethnic, social and neighbourhood safe grounds. Reflections about the relationship be- 1905structures and 1956. intact. Campbell et(1977) al. (2005) collected analysed ten these cases tween environmental conditions and mobility did- where relocation in Pacific islands happened between entistsnot start existed with thatclimate pollution change and and degradation its possible of imthe pacts. Already earlier agreement among social sci ten cases and identified another 76 communities in (Afifi 2011; Black et al. 2011; Lonergan 1998; Morris- the Pacific Islands region that experienced relocation (physical)sey 2013; Swain environment 1996; Wolpert can contribute to migration- between 1920 and 2004. Nearly 60 percent of these - cases related to environmental conditions, at least at 1966). Such common first sight. A closer look at prominent cases reveals,- place, however, often lacked in conceptual depth es- however, that often other aspects, e.g. economic and pecially in the questions of how such a relationship political-strategic considerations, were equally im actually looked like. The deterioration of environ portant for relocation, if not even paramount. mental quality or natural hazards can put people’s - McAdamwell-being, (2015) lives puts and livelihoodsit from “danger at risk zones”. to an extent International resettlement is more complex today that they move away from dangerous places or as than during colonial times as more actors are in- - volved. Colonial relocations often were internal, - within the boundaries of colonial possessions. Colo Since the notion of climate change and its possible im nial authorities could easily move people around in pacts on mobility has been widely discussed also his their empires. The colonial era has ended for most- e.g.torical Burson examples and Bedfordof environmentally-induced 2013; Campbell and mobilityBedford­ Pacific islanders. This does not mean, however, that 2014;in the ConnellPacific 2016,island 2015, region 2012; have Donner been mentioned 2015; Edwards (see the colonial experience is irrelevant for today’s chal 2014; McAdam 2014; McAdam and Ferris 2015; Weber lenges. Psycho-social aspects play an important role in decision-making around migration: fears about an - unknown future, fears of being isolated, deprived of- 2014b). The case study taken up for this paper is also- ernedthe protection by different that rules culture from and those citizenship people growprovide, up often shortly mentioned in an increasing body of litera fears of becoming powerless in a worldFritze which et al. 2008).is gov ture without going into details. The purpose of this pa - per is therefore to look closely at this case and assess in- with and which they understand ( - bertthe process Islands ifto there the Phoenix were only Islands environmental or if it is possible reasons to Such fears play an important role also among citi that brought hundreds of people from the southern Gil zens of Kiribati who face uncertainty about the fu- - ture of their atoll republic. Many are aware of these identify other layers of interests of the colonial power- historical cases of mobility as members of their fami as well as of the settlers. To answer these questions pri lies migrated­ in the colonial past. - mary material from colonial times has been used espe - cially communication and reports of decision-makers tledThe importancein 1938, then of again the PISS in the thus 1950s, goes and far abeyond third time his of the colonial power. These primary sources were not torical curiosity: Few people who were first reset- (availableBedford 1967;when Knudson the Phoenix 1965, Island 1977). Settlement Scheme PISS was scientifically assessed more than 40 years ago in 2007, are still alive. Still their stories, and the real 154 ity these stories pass on to moreDIE ERDE recent · Vol. generations, 147 · 2/2016 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

out in an over-hasty and unprepared manner. are powerful social constructs that influence and- scheme was carried out at all, and why it was carried even shape decisions and actions. People’s life paths Itgive helps insights to understand that are valuable the role naturein the conceptualiza plays (or does nottion play)of climate in determining change mobility human as behaviour; it emerges nature today. The idea of this paper originated during fieldwork about community resettlement on in- the Western Province of the after in form of climatic variability and changes as well as a tsunami that happened in 2007. Many of the par geological events such as earthquakes and tsunamis.- ticipants in interviews had come from the Phoenix - andIslands were in brought the 1950s to theor were Phoenix descendants Islands in of1938/39. people Kiribati provides an excellent example to under who originated from the southern stand how historical events continue to have an im- - portant meaning for the present and the future. Two- edIn by 2012, the 2007 and again tsunami in 2013,to set up the a authorneeds assessment conducted major resettlement events have shaped the coun fieldwork in the communities most severely affect try’s legacy with mobility: 1) the (forced) resettle- ment of the people of to Rabi Island in Fiji in some five years after the tsunami. The PISS was by December 1945, and 2) the (voluntary) Phoenix Is no means in the centre of the research objectives. land Settlement Scheme. The relocation of Banaban But issues concerning this settlement scheme came population to Rabi has been well documented. PISS- up during interviews again and again and caused a on the other hand is hardly documented; the major- thorough literature research on the topic. This then academic work was done before secret and confiden also brought confidential documents on the scheme tial material from colonial archives became avail- from the 1930s and 1940s to light, which provide videdable. This by Bedford paper is (1967) a first attemptand Knudson to provide (1965; a sketch1977). most of the primary sources for this paper. of an alternative perspective to what has been pro 2. The Republic of Kiribati

New primary sources make it necessary to alter the - characterization as well as the evaluation of this last- mentalBritish (meaning colonization land effort pressure in the as Pacificthe result Islands. of high In The Republic2 of Kiribati is a small developing is particular it has to be questioned whether environ land state in the2 Central Pacific. It has a land area- of 726 km (GoK 2010a, b) widely scattered across population density) was the major reason, why this 3.5 million km of the Pacific Ocean. The country con

1700 E 1750 E 1800 1750 W 1700 W 1650 W 1600 W 1550 W 1500 W N MARSHALL Kingman (USA) ISLANDS Palmyra (USA) 5N0 Washington () s I d l n s a i e L n Gi l b e r t I s l a n d s Fanning K I R I B A T I () Northern Marake Northern Christmas Howland (USA) () Baker (USA) Jarvis (USA) 0 Central 0 Banaba NAURU Southern Kanton (Abariringa) Tamana Enderbury McKean Birnie Phoenix () Malden Hull 5S0 () () () Starbuck P h o e n i x I s l a n d s Southern

TUVALU COOK Caroline SOLOMON ISLANDS 100 S ISLANDS Vostok Flint WALLIS and 0 250 500 750 mi VANUATU FIJI FUTUNA (French) 0 500 1000 km FRENCH Fig 1 The Republic of Kiribati

DIE ERDE · Vol. 147 · 2/2016 155 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

­(Mueller-Dombois and Fosberg sists of 32 islands concentrated in three island groups 1998: 632). This zone with a west-east extension of almost 5,000 km. In the extends westwards to the southern Gilbert Islands- very west there are 16 islands which form the Gilbert and southwards to northern . As a result of- andgroup; eight eight islands islands in situatedthe very some east of1,500 Kiribati km east form of the their flatness the lack substantial oro southern Gilbert Islands make up theFig. Phoenix 1 group, graphic rainfall. Rainfall variability in the Phoenix Is lands is extremely high. “These islandsMueller-Dombois can be without Northern and Southern Line group ( ). All these andany significantFosberg rain for several years in a row, while- islands are low-lying or islands. Banaba, in other years they can be very wet” ( 21an of individual the 33 islands island are west inhabited. of the By Gilbert mid-2015 Islands, Kiriba is a- most northern1998: (and 317). driest) Average of the rainfallPhoenix in Islands, the Phoe has raised with a maximum altitude of 82 m. annix average Islands increasesannual rainfall from northof 509 to mm south. and Kanton, a potential the ti had a population of 113,400 people (SPC 2016). The - country’s annual population growth is 2.2 % ranging evapotranspiration (PET) of 5620 mm, making this third among the Pacific Islandth countries (PICs)th globally behind island extremely dry. The more southern islands re the Solomon Islands (2.7 %) and Vanuatu (2.6 %) (SPC yearsceive more(Mueller-Dombois rainfall, but and also Fosberg here extended 1998: 29; droughts Thomas 2015). Kiribati ranks 12 of 14 PICs and 137 - happen rather frequently and can extend over several in the 2015 Human Development Index (UNDP 2015). Economic, environmental and social development in 2009). In this part of the Pacific Ocean El Niño events dicators such as GDP per capital, child morbidity and cause unusually warm water temperatures, which- infant mortality, access to water and sanitation are then result in higher rainfallObura than et during al. 2016). La Niña among the lowest in the Pacific (SPC 2015). events which bring cooler water temperatures con nected to drought conditions ( The Phoenix Islands, which play a major role in the 3. The Phoenix Island Settlement Scheme – the paper, consist of eight small low-lying coral islands political construction of overpopulation? islandsbetween is 2° 46’ and 4° 39’2 South and 170° 43’ and 174° 32’ West (GoK 2009). The total land area of the th about 29 km . The most northern island is until the mid-20th (Abariringa in ).- At first sight PISS resembles schemes of the later 19 - To the south there are , Rawaki century: to transfer populations (Phoenix Island), Manra (Sydney Island), Birnie Is- “from high-density ‘dangerMcAdam zones’ to low-densityBed ar- land, Nikumaroro (Gardner Island) and Orona (Hull fordeas”, (1967)to use andland Knudson more efficiently and avoid conflicts- Island). All except one island (Kanton Island, 31 per- over limited resources ( 2015: 95). Also - sons in 2010) are uninhabited (GoK 2010a). In 20062 the (1965) adopted such a Mal the Government of Kiribati created the Phoenix Is thusian deterministic approach puttingMaude’s great empha lands Protected Area (PIPA), with 408,250 km sis on the carrying capacity of a particular area. This alargest world Marineheritage Protected site, the largestArea (MPA) and deepest in the Pacific world approach was already dominant in thinking heritageOcean (GoK site 2009).in the world In 2010 (Toonen UNESCO et al. declared 2013). PIPA when arguing in favour of the PISS: People had to be 1931resettled Henry to “Harry” avert harm Evans happening Maude to them. After a census of the Gilbert, Ellice and Phoenix Islands in- - , who later became- nesians)There is evidencein pre-European that Manra, times Orona (Ellis-Jones and Nikumaroro 2014), but lengein charge of overpopulation of the settlement and scheme, hunger stressed in the southernthat “mi thewere populations inhabited byeither Polynesians died out (andor the possibly islands Micro were gration seemed toMaude be the obvious answer” to the chal - the beginning of the 19th landsGilbert were Islands brought ( to the1952: Phoenix 66). PISS Islands. started in 1938 abandoned long before the discovery by Europeans at when the first settlers from the southern Gilbert Is century. Two islands north of the Phoenix group, Howland and Baker Islands, are unincorporated territories of the USA. - The scheme was about the colonial object, the Gilbert - and Ellice Islands Colony, its uses (and particularly- The Phoenix Islands are located in the Central Pa extensions) and how to treat people living in colonial- cific dry zone characterized by “arid, desertlike cli spaces. Pacific islanders’ cultures looked strange, bar mates156 along the equator in the middle of the Pacific” baric, uncivilized, and – of courseDIE ERDE – inferior · Vol. 147 to · 2/2016 Euro Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

Maude, an anthropologist by profes- Maude and Maude - peansurely civilization. with humanitarian undertones, intentions and Fiji, and by 1881 ( - sion, wanted to bring civilization to the uncivilized, 1981). It seems that since the 1860s a consider able decline in population of the southern BedfordGilbert Iset methods; to the benefit of the people. He wanted to al.lands 1980, had Maudehappened. 1937a) In 1931 translating population into figures an average were save Pacific races from depopulation, by helping them much lower compared to earlier decades ( 2 peopleto keep who at least did notpart need of their to regulate cultural their identities population when 2 70 years converting them into Christianized colonial objects;- population density of aroundBedford 110 et inhabitants al. 1980 and per Maude km Maude compared to 220-250 inhabitants per km - with barbaric methods such as abortion and infanti nau)earlier population (calculated densities from in 1931 were around a third cide ( 1937a). His scheme in the end turned out of1937a). what Forhad individualbeen reported islands for (e.g.1861. Nonouti and Niku to become less humanitarian than intended. It finally- nialwas carriedadministration out to benefit led to strategicthe implementation colonial interests. of the Asymmetric power between the settlers and the colo 4. Humanitarian motifs vs. colonial economic - ­interest in the Phoenix and scheme, in particular asymmetric knowledge about- the physical environment of the islands where a con Maudesiderable part of the population of the southern Gil Maude (1938a) suggested four of the eight Phoenix bert Islands was taken to. There is much evidence that- Initially interest in the scheme was overwhelming. and other colonial administrators were aware of the extreme climatic variability of the Phoenix Is- Islands to receive settlers: Hull, Sydney, Gardner and- lands, but still the project was carried out in a haste Kanton Islands. Beneficiaries of the schemeKnudson should (1977) be and without safeguards to protect people from pro the poorest from the southern Gilbert Islands, espe- longed droughts. Still, as one can see further down,- cially those with no or little own land. the settlers surely displayed quite a lot of agency; it is remarks that some of the settlers actually had consid not possible to describe them as mere victims of colo erable land resources in the Gilbert Islands. Settlers nial power plays. A more realistic assessment is that took the opportunity offered to them. The risk was they first were not able to make informed decisions, limited: each settler – no matter if male or female, as crucial information was hidden from them, and adult or child – received a plot of land with at least 100 later, at least it seems, they tried to rectify this error. - providedbearing as well. trees. For a Alimited family time of four food thus rations was givenwere land with 400 fructiferous trees. Infrastructure was Whether the southern Gilbert Islands were overpopu up land rights in their home islands the land remained lated when the scheme was born is difficult to say. As withsent at their government’s family (Maude cost. Although settlers had to give much as the concept of carrying capacity is contested - from its conceptual side, it is also extremely difficult gration was nothing new 1937a).to people The from opportunity the southern was to define a correct number or density of people for a- tempting for people in densely populated atolls. Mi ulationparticular during location the 19 thatth and is early sustainable. 20th The southernBedford Gilbert Islands must have experienced a decline of pop Gilbert Islands: by the 1930s thousands had migrated century ( within the past 70 years. With becoming part of PISS et al. 1980). Wars between islands, diseases introduced- people extended the base of the households’ livelihood by Europeans, droughts and labour recruitment to Fiji, system by diversifying livelihood sources. - 19Samoa,th and , early 20th Hawaii and South and Central Ameri- ca had severe impacts on population numbers. The late- There is little doubtMaude’s that PISS also intended to colo- century was also a time when con lonialnize uninhabited government. islands In 1938, for e.g., economic Maude (colonial) presents again. de- cerns were expressed that PacificRivers societies 1922). would dis This is stressed in reports to the British co appear; many publications looked into the possibility of Pacific depopulation (see e.g. tailed budget on the estimated costs and revenue for- Hull and Sydney Islands. AccordingMaude to1937a). his calculation In the se- After theMcCreery turn of and the Munro century 1993; many Munro labourers 1992; Munro were the annual costs would be some £ 312 while the an- alsoand Firthrecruited 1987; for Rennie the phosphate 1987; Siegel mines on Banaba and nual revenue would be £ 619 ( Nauru ( Maudecret Report (1940) on theelaborates Phoenix inand great Line Islandsdetail about with Spethe 1985). Alone from cial Reference to the Question of British Sovereignty TabiteueaDIE ERDE · Vol. about 147 3,000 · 2/2016 labourers had left for Samoa, 157 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

- sideration of the provision of an outlet for surplus na- theeconomic British potential Empire. ofMaude’s 12 islands analyses, in the however, Phoenix were and had stressed that “in addition to the important con- Line groups and how these islands can be secured for tive population, the effective occupation of the Phoe offar the too 19 optimistic.th resources in the equatorial- nix Islands may become a matter of urgency to ensure islands had been grossly depleted in the second half their retention as British possessions” (AHC 1936: 1). century and copra prices were severely af The same letter provides some background why the- fected by the Great Depression. Even before the Great British considered the island group so important: Depression, however, colonial companies struggled to- “With the development of air transport and the con- profitably operate small isolated islands where high tionsequent of the increasing Phoenix interestIslands inrenders Pacific their air routes, retention from a transportation costs and low production volumes con- both commercial and defence points of view, the posi flicted. With the slump in copra prices and the start of wasWorld hardly War anythingII the settlers to gain on from the Phoenixthese islands. Islands be matter of considerable importance, and it is necessary came forgotten. Except for these two resources there that preparations for action should be made as soon as- missionerpossible to appointedsecure British Maude interests in the group” (AHC 5. Imperial power struggle – to boldly go where no 1936: 2). In November 1938 the Acting Resident Com- man has gone before? as officer-in-charge of PISS. “Owing to political considerations it is impera- tive that advance parties of natives should be landed at Gardner, Hull and Sydney Islands at the earliest op- theirBringing lives Gilbertese to even greater people dangers to the Phoenix than in Islandstheir home did portunity and you should make every endeavour to not take them away from danger zones, but exposed settlersexpedite left the for equipping the Phoenix and Islands.dispatch of the first expe - dition” (ARC 1938). Less than a month later the first islands. Scarcity of drinking water, frequent droughts, By that time the idea to bring settlers to Kanton Island and difficulties to even provide the very basis for sub sistenceMaude’s production were the reasons why the islands - enoughwere uninhabited safe water whenfor hundreds discovered of settlers by Europeans. (Maude had been given up, but not because it was realized that Also reconnaissance trip in 1937 did not find particularly this island was most unsuitable for colo nization because of erratic water supply. The British 1937a, 1937b); a reconnaissance trip which had to be Empire was in great and fierce competition with the suppliescut short and because water of were water giving scarcity. out” ( “WeMaude should 1952: have 77). about Kanton Island. In 1937 there even liked to have stayed longer in the Phoenix group, but (wasMegaw a confrontation between the British warship HMS - Wellington and the American warship USS Avocet 1977). In April 1939 Great Britain and the USA Stillsupply the Maude scheme was carried through although prep agreed to jointly administer the island for fifty years, arations were incomplete. To secure reliable water (andSamuels “thereafter until such time as it may be modified or- (1937a) had suggested the construction- terminated by mutual consent of the two governments”- of five cisterns. Less than a year later he revised the 2008: 740). During 1938 and 1939 Pan Ameri againstnumber ato possible four - iffailure possible in the to wellbuild water them. supply “The andcis can Airways deepened and cleared the , devel Iterns personally are only doubt intended if they to will be anever emergency have to be reserve used” oped an extensive airport and conducted flights to New (Maude Zealand using Kanton as a refuelling station. -

1938b: 2). MaudeThe cisterns 1952). were never built and The 1930s were a period of strong competition be already the first batch of settlers experienced severe tween the USA and Great Britain for several islands water challenges ( in the Central Pacific, including the Phoenix Islands. resettlement started; things at that time had to be The of 1856 entitled US-American Political considerations had become paramount when citizens to claim islands that had guano deposits for the USA, if these islands were uninhabited and not done quickly. The Phoenix Islands started to play a whalingunder the ships law inof theanother 19th country. Although most of strategic role for Western powers. The British saw it the Phoenix Islands had been discovered by British crucial to settle at least a few native people on these century the USA put claims islands permanently – at any cost. In November 1936 on a number of islands in the Central Pacific. In March 158the Acting High Commissioner for the Western Pacific 1935 the USA re-confirmedDIE their ERDE rights · Vol. 147 concerning · 2/2016 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

Jarvis Island and in May 1936 the same was done with Drought, isolation and World War II put a heavy toll regard to Baker and Howland Islands. The British on the settlers. When the war was over Britain had were worried that also islands in the Phoenix Group other concerns than to bother about a few people could get lost to the USA, if they were not under the on the Phoenix Islands. The strategic picture had jurisdiction of the High Commissioner of the Western changed entirely. The USA emerged as a superpower Pacific. It is interesting to note that around this time- exercising control over the Pacific, in both military the USA tried to establish a permanent presence on and economic terms. Kanton Island had become a some of the Central Pacific atolls. They started a Bak PANAM refuelling station already before the war and er, Howland and Jarvis Colonization Scheme which with the emergence of long-haul aircrafts became brought US-American settlers to these islands. The obsolete over the years. Copra production from the scheme was discontinued less than a decade later and Phoenix Islands had never been lucrative. It turned- the towns of Millersville (), Itascatown out to be far too costly to send once or twice in a year () and Meyerton () given a ship to the islands to collect copra. A long and se- Maudeup. Today these islands are uninhabited. vere drought between 1949 and 1952 put the scheme finally over the edge. The first to leave were the peo- wanted to take 750 GilberteseMaude to the Phoenix ple of Sydney (Manra) Island. Already in 1951 many Islands. His initial plan included Kanton Island, which demanded to be evacuated. This then happened be- was taken off the list (see above). calculated tween 1955 and 1957 and the people were taken to that once settlement and agricultural activities had the Protectorate. The popu started, the Phoenix Islands could carry up to 4,300 lation of Hull and McAdam Gardner (2014: Islands 304) arrived suggests there that in settlers (3,100 without Kanton Island). Locals who 1963-64. Based on interviews made on Tarawa, the had accompanied him on the reconnaissanceTable 1). trip saw andcapital “that of Kiribati,the Phoenix Islanders deliberately poured a potential of up to 1,400 which – over the years – the people’s concern might not have been genuine couldMaude reach almost 9,000 persons ( - - seawater into their wells so that when inspectors (1937a) proposed to carry out the scheme in came, they would believe that the water was not po- monthstwo stages: surveying to send and small sub-division working parties of land to proper the is- table, and this would support the islanders’ desire to lands and prepare them for settlement. Within five havemove”. been This to might get awayhave beenfrom thethese case, islands. showing Donner peo wells, planning the village sites, and generally pre- ple’s agency, showing how desperate people must ties, “constructing of cisterns, sinking of further reports of sentiments “that life was pleasant in the (Maude (2015), based on interviews on Ghizo Island, also paring the island for the settlers” could be completed - 1939 (Maude1937a: 23). The actual colonization should be- Phoenix Islands: ‘plenty to eat’ […] ‘many things in istration,carried out however, after the pushed 1938-39 heavily rainy toseason start inwith March the Sydney Island’”. Other reports, including own inter- 1937a: 24). The British colonial admin tratorsviews with to get Micronesians away from the on Phoenix Ghizo Islands Islands. suggest otherwise. People put pressure on colonial adminis scheme already in late 1938.

Table 1 Estimates for the Phoenix Island Resettlement Scheme, compiled from Maude (1937a, 1940); Knudson (1965, 1977); ­present population: GoK (2010a)

Estimation Estimation Estimation Estimation Final number Maude Maude locals locals Present of resettled population for initial for total for initial for total people settlement settlement settlement settlement

Hull

Sydney 350 1100 500-700 1300-2000 530 0 Gardner 400 900 600-700 1400-2200 300 0 Canton n.a. 1100 n.a. 1500-1600 97 0 total n.a. 1200 n.a. 3000 0 31 750 4300 1100-1400 7200-8800 927 31 DIE ERDE · Vol. 147 · 2/2016 159

Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

6. Resettlement of Phoenix islanders to the ­Solomon Islands - to their houses. The road connecting the settlement to - Gizo town is a challenge for four-wheel-drives and un Residentsusable for ordinaryof the new cars, settlement taxis, trucks assume and thatmini-buses. the ne- The Western District of the Solomon Island Protector ate was selected to receive settlers from Phoenix as it was sparsely populated. Ghizo Island e.g. had locations glect of their settlement is because the provincial with no traditional land tenure systems. Except for Gizo government wants them to move back to the coast. town only two Melanesian settlements were on Ghizo They are squatting on government land. Providing Island. The location for Titiana, the new MicronesianKnudson infrastructure would encourage them to stay where village, was identified some three kilometres west of they live right now. In the coastal villages agriculture Gizo town on the southern coast of the island ( is restricted because of sandy soils, but now they can 1977). In September 1955 some 30 men arrived from grow sufficient food for their families. Some even- Sydney Islands to prepare village and garden sites. Over started commercial agriculture, selling their produce- the next few years settlers came from the Phoenix and at the vegetable market at Gizo town. Very few re Gilbertese Islands. In 1958 the last and with 215 settlers member the time on the Gilbert Islands before reset the biggest group settled in New Madra near Titiana tlement to the Phoenix Islands, but many experienced- when the last people of Sydney Island were evacuated. the difficult time on the Phoenix Islands and consider Other settlements were created on totheir move new here settlement after the as tsunami the best and they they ever wish had. to Unani stay. (west of Ghizo Island) in the earlyBedford 1960s 1967). and on Wagina mously they agree that it had been the right decision- Island (east of Choiseul Island) in 1964 receiving people from Hull and Gardner Islands ( They managed to diversify the sources of their live- - lihoods. Many frequently walk down to the coast for A third resettlement became reality for people of- fishing, but with their new land they have taken up ag- ­Titiana and New Madra when a submarine earth- tunities.riculture Yetas afor complementary many the situation livelihood has improved option. The to quake with a magnitude of 8.1 caused a tsunami af difficult road conditions constrain these new oppor fecting coastal villages of Ghizo, Choiseul, New Geor gia and other islands on April 2, 2007. The biggest such an extent that some see the tsunami that brought 33damage died inwas the along event the (Fisher southern et al. coast 2007); of 31Ghizo were Island. from them to “their” new land a ‘blessing in disguise’. They The death toll was 52 people. Alone on Ghizo Island only hope that they will not be forced back to the coast. 7. Conclusion the Micronesian community which had less than ten percent of Ghizo Island’s population. 21 victims of the tsunami on Ghizo Island were children under the age - Fiveof ten years years after (Solomon the tsunami Islands the 1 April wounds Tsunami of the 2007). event Mobility has social and cultural aspects to consider – - the social, economic and cultural structures of soci eties, which often connect the presence to a colonial- were still visible when research was carried out in Ti past and an imaginative future. This is also true for- tiana and New Madra in 2011 and 2012 to learn how- people’s perception and awareness of mobility. Posi people had recovered from the tsunami. Interviews tive as well as negative experiences shape expecta- were also conducted in settlements some three kilo tions and fears. Both have consequences for people’s peoplemetres builtinland permanent to where housesmany of on the higher Micronesians ground away had decision making. The Republic of Kiribati is an excel- evacuated themselves. Immediately after the tsunami lent example to show how historical events continue to have important meaning for the presence and fu from the coast. Some have returned to the coast, but- ture. People of the southern Gilbert Islands were not the majority today lives in the settlement called “Three- taken away from dangerous places, but put directly in- Miles”. Conditions in the settlement are harsh. Author such places to satisfy economic and political interests ities did not provide any infrastructure. The rehabili of their colonial masters. This contradicts a concep- tation process is incomplete. The most essential items tualization of mobility and environmental change, missing are secure water supply and proper roads where natural and human systems are coupled pro connecting the settlement and Gizo town. People have ducing a particular human agency: to move away from to walk long distances to take their daily bath in small dangerous places voluntarily or being taken away, 160creeks, wash their dishes and bring water in buckets resettled, relocated. This standsDIE ERDE in · Vol. gross 147 contrast · 2/2016 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

with the experience of people being “trapped” where interviewed have taken part in the entire odyssey: they are, meaning the reduction or even elimination brought as small children from the Gilbert Islands to thatof agency, see people or – atresponding the other toextreme environmental – people threats,moving the Phoenix Islands, then to the Solomon Islands some into dangerous places acting contrary to approaches 3,000 kms away to finally squatter in the hinterland - of Ghizo Island when their houses were washed away- risks and stress. For all these actions reasons other by tsunami waves. Climatic and ecological features than being driven by environmental factor play a de have played important roles in the lives of Microne cisive role. It appears that environmental aspects play (andsian peopleat times interviewed opportunities) on the Ghizo natural Island, environment their final least of a role and are insufficient to explain people’s station for the time being. To ignore these challenges action – surely not in the sense that people are driven by environmental conditions and changes. provides to people would be inappropriate. However, to the Phoenix Islands for no valid environmental rea- people’s mobility was never determined by natural People of the southern Gilbert Islands were resettled- forces, not even in cases of extreme environmental- and climatic stress and pressure. People always made- son. The places they lived before were environmen- use of their agency, were willing to respond to oppor tally more suitable than the places they were taken to. tunities and were slowed down and hindered by con yearsStill they before did theynot bother were given as they permission received anto leaveopportu the straints; still trying, still not powerless victims, but nity to diversify their livelihoods. It took a couple of agents that always tried to influence their destinies. - ‘dangerous places’ in the Phoenix Islands. After the- To give too much importance to colonial settlement tsunami on Ghizo Island they moved to the best place schemes might appear unjustified as during colo they experienced in the past 77 years or so. Well pro nial times it was much easier to move people around- tected against the dangers of tsunamis, endowed with like on a chess board. This situation has considerably fertile land they can grow a big variety of crops for changed. However, what remains is asymmetric pow their own use and for the market. Still plans exist to er relations. No matter, if environmental and climate- take them back to the ‘danger zone’ of the coastal area. change issues are included as valid reasons in refugee conventions and other relevant legislation, this asym Colonial (re)settlement schemes shaped the perception metric power suggests that the legislation is often not- of I-Kiribati about mobility. Their colonial past is less- even worth the paper it is written on. What is needed than 40 years ago. The removal of the Banaba people therefore is not legislative change, but change in po lifeby British span. In authorities 2004, when is thejust author 70 years had ago. the Even opportunity the set litical will and awareness. Otherwise it will inevitably tlement ofAnote the Phoenix Tong Islands falls within a person’s happen that making people climate change refugees will compromise their right to live in dignity. This to talk to , President of Kiribati, he became asymmetric power also becomes visible in recent trade- aware of the fact that events like colonial resettlement negotiations where Pacific Island countries demand shape the perception and the awareness of I-Kiribati in that labour mobility should become part of agree the present discourse about environmental mobility. ments with Australia and (PACER-Plus) Having taught many students from Kiribati over the as well as the European Union (­EconomicWeber 2015a). Partnership years this was confirmed: Today’s perception is closely Agreement). In both instances the PICs have so far not linked to the experiences of relatives and neighbours. been not able to achieve this goal ( - In academic discourses such historical events cannot be ignored or downplayed: For people’s decisions they are plausibleRight now that we thosedo not who even are know able theto move number – those of (addi with as important as the degree of environmental change.Weber tional) migrants as a result of climate change. It seems Other important aspects are Australia’s ill-treatment- of asylum seekers on Manus Island and Nauru ( money, social networks and alternative livelihoods – wants2015b). to Keeping attain a suchstatus historical of environmental as well recentrefugee. narra will migrate independently. The vulnerable poor,- tives in mind it becomes clear that nobody possibly those with little capacity to move when environments deteriorate, are possibly left behind or forced to reset tle later. They are those who might become climate-- afterThe paper a tsunami looked in into2007. what Few started people aswho PISS have in 1938been change refugees: with few capacities and skills, little and became an example of community displacement wanted by those countries they try to get to, depend DIE ERDE · Vol. 147 · 2/2016 ent on support and benevolence, and endangered161 to Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

Connell, J. - the possibility of people being left behind in inhumane 2015: Vulnerable islands: climate change, tec become a pawn in political in-fighting. There Kelmanis even 27 (1): 1-36 tonic change, and changing livelihoods in the Western Connell, J. environments. “Evacuating sinking islands” ( ­Pacific. – The Contemporary Pacific 2016: Last days in the Carteret Islands? Climate 2008) in the Pacific will most likely not happen. “It is 57 (1): 3-15 change, livelihoods and migration on coral atolls. – Asia more likely that the last plane and ship leaves Kiribati Donner, S.D. Pacific Viewpoint nor(or Tuvalu) environmentally” and people (areWeber still 2015a: there, not19). forgotten, but - never intended to be brought to safety, neither socially 2015: The legacy of migration in response to climate stress: learning from the Gilbertese resettle 39 (3-4): 191-201 ment in the Solomon Islands. – Natural Resources Forum Bibliography Edwards, J.B.

2014: Phosphate mining and the relocation of the Banabans to northern Fiji in 1945: lessons for climate 138-139: 121-136 Acting High Commissioner (AHC) for the Western Pacific change-forced displacement. – Journal de la Société des Ellis-Jones, I. 1936: Telegram [secret] to the Resident Commissioner, Océanistes th - Gilbert and Ellice Islands Colony, November 26th, 1936. – 2014: The Phoenix Islands – An annotated - Accessed online: https://digital.library. adelaide.edu.au/ chronology. – 4 edition. – Turramurra. – Accessed on - dspace/handle/2440/79867, 25/02/2015 line: http://de.slideshare.net/ianellis-jones/the-phoe Fisher, M.A., E.L. Geist, R. Sliter, F.L. Wong, C. Reiss and Acting Residential Commissioner (ARC) for the Western Pa nix-islands-an-annotated-chronology, 12/05/2016 D.M. Mann cific 1938: Letter to H.E. Maude, November 25th, 1938. – - Accessed online: https://digital.library.adelaide.edu.au/ 2007: Preliminary analysis of the earthquake Afifi, T. dspace/handle/2440/79903, 25/02/2015 (MW 8.1) and tsunami of April 1, 2007, in the Solomon Is 49, 26 (1): 3-20 2011: Economic or environmental migration? The lands, southwestern Pacific Ocean. – Science of Tsunami Fritze, J.G., G.A. Blashki, S. Burke and J. Wiseman push factors in Niger. – International Migration Hazards Barnett, J. and W.N. Adger - ­Supplement 1: e95-e124 2008: Hope, 61 (3): 321-337 motion of mental health and wellbeing. – International 2003: Climate dangers and atoll despair and transformation: climate change and the pro Bedford, R.D. 2 (1): 13 countries. – Climatic Change - 1967: Resettlement: solution to economic and Journal of Mental Health Systems social problems in the Gilbert and Ellice Islands Colony. – Government of Kiribati (GoK) 2009: Phoenix Island Pro Bedford, R., B. Macdonald and D. Munro 1980: Population ­estimates M.A. thesis in Geography, University of Auckland tected Area – nomination for a world heritage site. – - Bike­nibeu, Tarawa, Kiribati 89 (2): 199-246 for Kiribati and Tuvalu, 1850-1900: review and speculation. – Government of Kiribati (GoK) 2010a: 2010 census of popu Black, R., W.N. Adger, N.W. Arnell, S. Dercon, A. Geddes and - The Journal of the Polynesian Society lation and housing. Vol 1: Basic Information and Tables. – D. Thomas National Statistics Office. Ministry of Finance and Eco 21, 2: 2011: The effect of environmental change on nomic Planning. – , Tarawa human migration. – Global Environmental Change Government of Kiribati (GoK) 2010b: 2010 Census. Vol. Burson, B. and R. Bedford - Supplement 1: 3-11 Analytical Report. National Statistics Office and the SPC 2013: Clusters and hubs: toward a re Statistics for Development Programme. – Noumea, New Kelman, I. gional architecture for voluntary adaptive migration in the Caledonia Campbell, J. and R. Bedford ­Review 31: 20-21 Pacific. – The Nansen Initiative, Technical Paper. – Geneva 2008: Island evacuation. – Forced Migration Piguet, E. and F. Laczko (eds.): Knudson, K.E. 2014: Migration and climate change in . – In: 1965: Titiana, a Gilbertese community in the People on the move in a changing climate: the regional Solomon Islands. – Master thesis, University of Oregon, 2 Knudson, K.E. impact of environmental change on migration. – Global Department of Anthropology. – Eugene Campbell, J.R., M. Goldsmith and K. Koshy - Migration Issues . – Dordrecht et al.: 177-204 1977: Sydney Island, Titiana and Kamaleai: lands. – In: Lieber, M.D. (ed.) 1977: Exiles and migrants 2005: ­Community southern Gilbertese in the Phoenix and Solomon Is - 5 relocation as an option for adaptation to the effects of Lieber, M.D. climate change and climate variability in Pacific is in Oceania. – ASAO monographs . – : 195-241 5 land countries (PICs). – Asia-Pacific Network for Global (ed.) 1977: Exiles and migrants in Oceania. – Connell, J. Lonergan, S. Change Research. – Tokyo ASAO monographs . – Honolulu - 2012: Population resettlement in the Pacific: 1998: The role of environmental degradation in pher 43 (2): 127-142 4: 5-15 ­lessons from a hazardous history? – Australian Geogra population displacement. – Environmental Change and Security Project Report 162 DIE ERDE · Vol. 147 · 2/2016 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

Maude, H.E. - McDowell, G., J. Ford and J. Jones

1937a: Report on the colonization of the Phoe 2016: Community-level - nix Islands by the surplus population of the Gilbert and ­climate change vulnerability research: trends, progress, 11 (3): 033001 Ellice Islands. [Confidential]. – Suva. – Available: Univer and future directions. – Environmental Research Letters online: http://hdl.handle.net/2440/79970, 25/02/2015 Megaw, M.R. sity of Adelaide, H.E. Maude Digital Archive. – Accessed Maude, H.E. 1937b: Report of Phoenix Islands ... vis- 17 1977: The scramble for the Pacific: anglo‐­ (69): 458-473 united states rivalry in the 1930s. – Historical Studies - Morrissey, J.W. ited in H.M.S. Leith. 28 May 1937. – Accessed online: https://digital.library.adelaide.edu.au/dspace/bit 2013: Understanding the relationship between - stream/2440/79869/1/Report%20of%20Pheonix%20 environmental change and migration: the development of 23 (6): 1501-1510 Islands...%20visited%20in%20H.M.S.%20Leith.%20 an effects framework based on the case of northern Ethio Maude, H.E. - Mueller-Dombois, D. and F.R. Fosberg 28%20May%201937.pdf, 25/02/2015 pia. – Global Environmental Change port on the proposal to alleviate over-population and 1938a: Gilbert and Ellice Islands Colony. Re 1998: Vegetation of the - and synthesis 132 tropical Pacific islands. – Ecological studies. Analysis Munro, D. “land-hunger” in the colony. [Confidential]. – Ocean Is . – New York et al. land. – Available: University of Adelaide, H.E. Maude 1992: Gilbert and Ellice islanders on Queensland net/2440/79906, 25/02/2015 14 (11): 449-465 Digital Archive. – Accessed online: http://hdl.handle. canefields, 1894-1899. – Journal of the Royal Historical Maude, H.E. Munro, D. and S. Firth Society of Queensland - 1938b: Memorandum. Local steps to be taken 1987: From company rule to consular - - if and when approval of the Phoenix Islands Settle control: Gilbert island labourers on German plantations 16 (1): 24-44 ment Scheme is received. [Confidential]. – Ocean Is in Samoa, 1867-96. – The Journal of Imperial and Com Obura, D., S. Donner, S. Walsh, S. Mangubhai and R. Rotjan land. – Available: University of Adelaide, H.E. Maude monwealth History net/2440/79918, 25/02/2015 Digital Archive. – Accessed online: http://hdl.handle. Maude, H.E. vulnerability assessment and management. Report to 2016: Phoenix Islands Protected Area climate change - - 1940: Report on the Phoenix and Line Islands with Special Reference to the Question of British Sov the Aquarium, Boston. – Updated. – Ac ereignty. [Secret] – Suva. – Available: University of Ade­ cessed online: http://www.phoenixislands.org/pdf/ http://hdl.handle.net/2440/88870, 25/02/2015 Rennie, S. - laide, H.E. Maude Digital Archive. – Accessed online: PIPA-CC-scoping-study-Jan-18-2016.pdf, 14/05/2016 Maude, H.E. 1987: Contract labor under a protector: the Gil - 11 (1): 81-106 1942: Memorandum. The inception of the bertese laborers and Hiram Bingham, Jr., in Hawaii, Rivers, W.H.R. (ed.) 1922: Essays on the depopulation of scheme for colonizing the Central Pacific Islands. Febru 1878-1903. – Pacific Studies ary 12th, 1942. – Available: University of Adelaide, H.E. handle.net/2440/79870, 25/02/2015 Samuels, J.H. - Maude Digital Archive. – Accessed online: http://hdl. Mela­nesia. – Cambridge Maude, H.E. 2008: Condominium arrangements in inter 61 (1 & 2): 62-89 - 1952: The colonization of the Phoenix Islands. – national practice: reviving an abandoned concept of Maude, H.E. and H.C. Maude 29 (4): 727-776 Journal of the Polynesian Society boundary dispute resolution. – Michigan Journal of In 1981: Tioba and the Tabiteuean ternational Law 90 (3): 307-336 religious wars. – The Journal of the Polynesian Society Secretariat of the Pacific Community (SPC) 2016: Statistics McAdam, J. - for Development Division (SDD) website. 2015 Pocket 2014: Historical cross-border relocations in the Statistical Summary of Pacific Island countries and ter 49 (3): Pacific: lessons for planned relocations in the context of ritories. – Accessed online: http://sdd.spc.int/images/ 301-327 climate change. – The Journal of Pacific History documents/Pocket_Summary/2015_Pocket-Statistical- McAdam, J. - Summary.pdf , 07/06/2016 2015: Relocation and resettlement from colonisa Secretariat of the Pacific Community (SPC) 2015: National 3 (1): 93-130 tion to climate change: the perennial solution to ‘danger Minimum Development Indicator Database. – Accessed McAdam, J. and E. Ferris Siegel, J. zones’. – Review of International Law online: http://www.spc.int/nmdi/, 07/06/2016 - 20 (1): 42-54 2015: Planned relocations in the 1985: Origins of Pacific Islands labourers in Fiji. – Swain, A. - context of climate change: unpacking the legal and con The Journal of Pacific History 4 (1): 137-166 ceptual issues. – Cambridge Journal of International and 1996: Environmental migration and conflict dy McCreery, D. and D. Munro - 17 (5): 959-973 Comparative Law namics: focus on developing regions. – Third World Thomas, F.R. 1993: The cargo of the Montser Quarterly 49 (3): 271-295 63 (4): 567-600 rat: Gilbertese labor in Guatemalan coffee, 1890-1908. – 2009: Historical ecology in Kiribati: linking The Americas past with present. – Pacific Science DIE ERDE · Vol. 147 · 2/2016 163 Only a pawn in their games? environmental (?) migration in Kiribati – past, present and future

Toonen, R.J., T.A. Wilhelm, S.M. Maxwell, D. Wagner, B.W. Bowen, C.R.C. Sheppard, S.M. Taei, T. Teroroko, R. Moffitt, 3 (eds.): A new perspective on human mobility in the South. – C.F. Gaymer, L. Morgan, N. Lewis, A.L.S. Sheppard, J. Parks Weber, E. Global Migration Issues . – Dordrecht et al.: 119-148 and A.M. Friedlander to resettle. – In: Andrianos, L., J.-W. Sneep, G. Kerber and 2014b: Of tsunamis and climate change: the need - 2013: One size does not fit all: the 77 (1): 7-10 emerging frontier in large-scale marine conservation. – R. Attfield (eds.): Sustainable alternatives for poverty re - Marine Pollution Bulletin duction and eco-justice, Vol. 1. – 2nd edition. – Newcastle Weber, E. Solomon Islands 1 April Tsunami 2007. – Tsunami Newslet upon Tyne: 192-208 - ter 39 (2). April-June 2007: 1. – Accessed online: http:// 2015a: Envisioning South-South relations in the - www.pacificdisaster.net/pdnadmin/data/original/ fields of environmental change and migration in the Pa 2 (1): 6 ITIC_2007_Apr_Jun.pdf, 07/07/2015 cific Islands – past, present and futures. – Bandung: Jour Weber, E. United Nations Development Programme (UNDP) 2015: nal of the Global South 3 (4): 96-107 ­Human development report 2015: work for human 2015b: The Pacific solution – a catastrophe for the Weber, E. Wolpert, J. - ­development. – New York Pacific!? – Environment and Ecology Research Anich, R., J. Crush, S. Melde and J.O. Oucho 22 (4): 92-102 2014a: Environmental change and (im)mobility in 1966: Migration as an adjustment to environ the South. – In: mental stress. – Journal of Social Issues

164 DIE ERDE · Vol. 147 · 2/2016