+ All Categories
Home > Documents > Economy Militery Pre Islamic Luwu-18_04_Bulbeck

Economy Militery Pre Islamic Luwu-18_04_Bulbeck

Date post: 04-Jun-2018
Category:
Upload: lupa-ingat
View: 221 times
Download: 0 times
Share this document with a friend

of 14

Transcript
  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    1/14

    Al i TRALASIA HISTORlCALARCHAEOLOGY, 18,2000

    Economy Military n Ideology in Pre Islamic Luwu SouthSulawesi Indonesia

    F DAVID BULBECK

    A slim body indigenous texts, comprising lists rulers, an emic political geography and variousapocryphal tales, provides essential clues to the nature the Bugis chiefdom Ware Luwu) leading up toits adoption Islam in 1605. However, to appreciate Luwu economic history and evolving settlementpatterns, recourse must be made to the archaeological record. The resulting data are intelpreted in light the distribution Luwu main resources and its numerous language groups. High-grade iron ore the mostimportant commercial resource, was smelted locally and traded rexotic goods by thefirst millennium AD.Between c. I300 and I 6 Bugis settlers occupied critical coastal locations and made Malangke into theregion emporium. Luwu Bugis elite deftly combined control over the iron industry with deploymentnaval and ground troops, and elaboration a partly mystical, partly pragmatic ideology, to transform Luwutemporarily into the most prominent chiefdom in South Sulawesi.

    BACKGROU D TO LUWU AND ITSARCHAEOLOGICAL STU Y

    polities in the peninsula. Th e Bugis script had been developedin the fourteenth century, and there are several relevant Bugist ex ts to a ssist t he p re se nt study. T he se i nc lu de l ists of the preIslamic rulers of War e , t re at ie s w it h o th er Bug is c hi ef do ms,and a text that lays o ut L uw u s political structure. However,t he h is tori ca l r ec or d on Luwu is so sparse that no sourceexternal to Luwu names any of its pre-sixteenth century rulersa nd , unti l 1992, s ch ol ar s were oblivious o f the pre-Islamiccapital at Malangke. Ware clearly became a mighty chiefdom,bu t the illumination of its s oci al a nd economic organizationdepends on the ar cha eo logi ca l record. 4 Th e Or ig ins ofComplex Society in South Sulawesi or O XI S Pro je ct , w hi chi nv ol ve d sur ve y a nd e xc av at io ns in Luwu in 1 99 8 a nd 19 99,provides the archaeological evidence utilised in this article.

    Until the year 2000, when Luwu was divided into twodepartments, it covered a vast area at the northeast extreme oft he p ro vi nc e o f South Sulawesi see Fig. I). Two mainlandfonns c an b e i de nt if ie d, a c oa st al p la in t ha t m ay e xt en d u pto 50 rn inland, and a m ou nt ai no us i nt er io r w it h p ea ks t ha tfrequently exceed 2 000 m in altitude. Annual rainfallg en er al ly r an ge s b et we en 2 5 00 an d 3 000 m m, a nd vari esfrom slightly seasonal south of Palopo to pernlanently humidfarther north. Th e hot equatorial climate o f the coastal plaina nd f oot hi ll s g iv es w ay to cool d ay s an d c hi ll y n ig ht s a t h ig haltitudes. Th e great majority o f the population congregatesalong the coastal strip, although substantial settlements havealso built up al ong the less dissected reaches at middlinga lt it ud es , s uc h as Lake Matano. Bugis is only on e a mon gLuwu s t en i nd ig en ou s l an gu ag es, w ho se d iv er si ty r ef le ct sboth the barriers to communication and an evident multiplicityo f distinct colonisation episodes.

    Bugis belongs to the S ou th S ul aw es i stock whichessentially comprises the indigenous languages o f Sulawesi ssouthwest peninsula. Th e spatial distribution o f t he e le ve ndialects o f B ugi s c los el y relates to the va ri ous provinci ald ep ar tm en ts w hi ch in turn, correspond to the Bugischiefdoms t ha t t he D ut ch i nc or po ra te d w it hi n t he ir c ol on ia lgovernment about a c en tu ry a go. Luwu enjoys the reputationo f b ei ng t he o ld est a nd m ost p re st ig io us Bugis chiefdom, butit is a ls o t he l eas t typi cal . Th e other Bugis chiefdoms e.g.Wajo a nd B on e) f lour is he d in t he ce nt ral t wo- th ir ds o f thepeninsula where few of t he i nh ab it an ts a re n ot Bug is. By atl ea st A .D . 1 40 0 t he se w er e a gr ar ia n c hi ef do ms, b ased on theintensive cultivation of m on so on -f ed w et rice. In Lu wu, by

    TRODUCTIONTim Earle s recent book on the paths that may lead to theformation of c om pl ex p ol it ie s t ak es a h isto ri ca l p er sp ec ti vet ha t r ec og ni se s t he r ol es of individual ambitions and interest_ oups. Ea rle a nal ys es the s trategies by w hi ch an e me rgi ng

    lite can exe rt cont rol over t he t hr ee ma in i nt er de pe nd en tources of power - the economy, military force, and a

    legitimising ideology. He illustrates hi s a pp ro ac h w it hp ro to hi st or ic e xa m pl es o f p re st ig e- go od c hi ef do msDenmark), staple-fmance chiefdoms Hawaii), and hill-fort

    hiefdoms Per u) . T hi s p ap er a pp li es Earle s structure ofa na ly si s to t he a nc ie nt B ug is c hi ef do m o f Ware Luwu) in

    ulawesi, the large orchid-shaped island between Borneo andt he Moluccas. Wa re was the name t he r ul in g Bugis elitea ppl ie d to t he poli ty, w hi le t he t er m L uw u is appropriate forth governed realm, including the political structure headed byWare . The results show that Ware /Luwu c ou ld b e l ab el le d a. tapIe-good chiefdom as it combined aspects o f Earle staple-finance and prestige-good chiefdoms.

    A cording to Earle, the eli te o f prestige-good chiefdoms t he c ol le ct io n a nd exchange o f highly valuable local

    ur es. s uc h a s a mb er, to d om in at e t he r ec ip roc al t ra de in h ig h- st at us g oo ds . Bennet Bronson envisagedi ely this specialisation in the long-distance trade ofr tigious goods when he formulated his model o f the

    outheast Asian dendritic economy. Bronson s economicunit are the large river systems where produce from t herugged hinterland is t ra de d d ow nstr ea m f or i mp or te d g oo dsrought upstream. Critical nodes in t he n et wo rk i nc lu de t he- oU ec ti ng c en tr e at t he ma in c on ve rg en ce o f t he h ig hl an d

    butaries, an d the coastal centre at the river s mouth.Bron on s m od el w ou ld appear b ro ad ly a pp li ca bl e to L uw u, h its r es ou rc e- ri ch h in te rl an d, a nd its c on ce nt ra ti on o f

    IC between hinterland and coast to a few main routes.. ~ e v e r as we s hall s ee , Luwu e ff ec ti ve ly r ev er se d t hemitional mode of prestige-good chiefdoms. Th e latter

    ne d splendid trifles to obtain goods locally perceived asrpensable many of them manufactured, such as weapons.

    . on the other hand monopolised an essential commodity- ~ a r e s o f high-quality iron - to acquire a suite of goodspr dominantly superfluous to the necessities o f life.

    Th e p er iod u nd er s tu dy e xt en ds to t he e ar ly s ev en te en thnrury. when W are s hifted its capit al from Ma la ngke toP al opo ee Fig. 1). It ma y also be cal led the pre -Is lami c

    riod because at around the same time, in 1605, Luwue mb ra e d Islam, shortly followed by all the other major

    3

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    2/14

    o 200 kms J

    3i OO Sulf of Bone

    o Excavated early historical sites1ron ore sources on the Rongkong River

    NOTE: Marano includes Rahampu'u I,Pandai Besi and Lemogola120i30 E 121 iOO E

    50,

    N

    to,

    Bukit .\ Pangiwangen i .0-\ , ,\ -.j \

    li) ./(\\

    Fig I: Study area of the OXISproject in UlVU department behveen the dot-dash lines , and its spatial relationship to thenorthern Bugis agrarian chiefdoms Inset, their locations numbered I to 4

    contrast, Bugis are numerically dominant solely along thecoastal str ip as far north as Malangke. Rice underpinned thesubsistence economy only in Luwu's southernmost extension,the 70 km long stretch of coastal plain south of Palopo.Elsewhere in the Luwu lowlands, sago was traditionally themajor carbohydrate staple. These points suggest that theWare' polity was established in direct association with thenorthward expansion of Bugis speakers, from their homelandin the southwest peninsula, into an unfamiliar environmentinhabited by other language groups. The magnet appears tohave been the iron reserves in Luwu's hinterland.

    n 1992 and 1994, Ian Caldwell surveyed the uplandlocations in Luwu that have a history of smelting iron, and thecomplex of pre-Islamic sites at Malangke.7 n 1997, the aXproject surveyed these and other localities to plan its scheduleof excavations. 8 Subsequently, the iron ore sources at BukitPorreo' and Bukit Pangiwangen (see Fig. I) could not befurther investigated as the heavy rains in 1998 rendered theaccess road impassable. Our attention focused on LakeMatano, which lies in an iron-rich ultrabasic fonnation, andthe Luwu coastal plain.

    Approximately 80 sites have been excavated or surveyed,producing a large body of material the analysis of which isstill at an early stage. One metre test pits were universallyemployed, except at the waterlogged Utti Batue site where anine square metre test pit was excavated to prevent the baulksfrom collapsing. The number of test pits varied between oneor two at the smaller sites, to a stratified random sample of test pits at the largest site ofPattimang Tua (5.3 hectares), anda systematic sample of 22 test pits at Pinanto (0.6 hectares).The main site types were smelting sites at Lake Matano;industrial sites, along the coastal plain, where imported ironwas worked; other lowland settlements involved in Luwu'strade; and six pre-Islamic burial grounds. The excavation unitswere features and stratigraphic layers, with layers over 5 cmthick exhumed in 5 cm spits. Radiocarbon determinations

    have been obtained from all contexts considered relevant tothe research problems and not adequately dated by theassociated artifacts. Other details of excavation methodologyare provided e lsewhere .9 The overriding goals of theexcavations were to sample the major axes in pre-IslamicLuwu's coastal-hinterland trade, and to gain a diachronicperspective on the long-term development of the economicsystem.

    Only the dating evidence relevant to the period underdiscussion, beginning at about 2000 years ago and lasting tillthe seventeenth century, is considered here. The AustralianNational University (ANU) radiocarbon dates areconventional dates, and the OZD, OZE (Lucas Heights,Sydney) and Wk (University ofWaikato) dates are acceleratormass spectrometry (AMS) determinations. These dates arecalibrated with the CALIB 3.03 computer program O andexpressed in tenns of the 95.4 confidence intervalrepresented by two standard deviations (Table I).Unfortunately, the late fifteenth, sixteenth and earlyseventeenth centuries are virtually indistinguishable on theradiocarbon calibration curve, yet these centuries correspondprecisely to the period when Ware' seems to have risen toprominence. This problem is largely offset by the largenumbers of high-fired ceramics, made in China, Vietnam andThailand, imported to Luwu between the thirteenth andseventeenth centuries. They facilitate the precise dating ofsites belonging to that period.L NGU GE ND ETHNI ITY IN SOUTHSUL WESIThe study of pre-Islamic Luwu requires a diachronic view onthe dialectic between its Bugis and non-Bugis speakers. As afirst step, we should summarise the ethnographic perspectiveon South Sulawesi. Social anthropologists working in the areatreat language and ethnic group as synonymous. With fewexceptions, South Sulawesi indigenes throughout the

    4

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    3/14

    tbe mass convers ion to Is lam during the six years between1605 and 1611. All the major South Sulawesi chiefdomsofficially embraced Islam, requir ing only a modicum offorceful persuasion on the part of those promulgating the faith.Islam was then promoted through a loosely coordinatedprocedure within each realm.

    Tbe increasing trend towards a political hierarchy has hadto be balanced against safeguarding tbe mutual interests of themajor constituents within the realm, lest insurrectiondestabilise the chiefdom s local structure. Fo r instance, localtraditions emphatically proclaim the status of Baebunta andWotu as the two major players, after Ware , in Luwu ssociopoli tical organisat ion within the area from Malangkenorthwards (see Fig. I). Altbough Baebunta and Wotu wereinhabited respectively by Lemolang and Wotu speakers, theyfollowed tbe same mortuary practice as the Bugis in Malangke- cremation o f the dead - in the centuries leading up to theirvirtually simultaneous conversion to Islam. Along the coastalplain eas t of Wotu, reliable evidence for cremations isconspicuous by its absence. 7 These points strongly suggestthat Luwu s dominant language groups employed the rite o fcremation to widen the gap between themselves and therealm s minor constituents. The apparent lack of pre-Islamiccremations at Tampinna and sites directly to the east impliesfew if any Bugis had resided there, even if the cremations atthe Baebunta sites and Wotu do not imply these were Bugisstrongholds, because mortuary practices can cross languageboundaries under appropriate circumstances.

    On e phenomenon o f particular relevance to understandinglanguage and ethnici ty in Luwu is tbe ethnographical lyrecorded continuing encroachment of Bugis across the bordersof its distribution. Especially where Islam prevails on bothsides of the boundary, Bugis families move into non-Bugisterrain, and tbe offspring of non-Bugis speakers often switchto Bugis. Fo r instance, in Luwu, the Lemolang (Baebunta)community is gradually abandoning its own language forBugis, despite proud memories o f a non-Bugis heritage. IS AtTurun Bajo the settlement nearest to Tampinna, theinhabitants claim descent from the Bajau sea gypsies yetemploy Bugis as their first language.9 The fecundity o f theBugis, and the opportunities that arise for individuals who linkup with Bugis networks, would appear to account for thisapparently spontaneous expansion of the language. Theremarkable propensity of the Bugis language to expand at itsmargins st rongly suggests that Bugis has never been moreprevalent in Luwu than at the present - all of Luwu, beyondtbe coastal strip extending to Malangke, has a fundamentallynon-Bugis past.

    The Bugis addit ionally have a tradition o f spontaneousmigration to establish remote enclaves modelled along tbelines of extended families, including the creation of fictitiousgenealogical links. At Lake Lindu in central Sulawesi, Bugiscoloni st s have wi thin two gene rat ions legitimised theirpresence by creat ing an elaborate spir itual landscape andredefining the indigenous culture hero as Bugis. o The LakeLindu case documents the creation of attachment to a place asone of the essential ingredients in local ethnic identity, alongwith the colonists language (Bugis) and religion (Islam, asopposed to the indigenes Christianity). A similar developmentmay have occurred at Cerekang (near Katue) and Ussu wherean enclave of Bugis speakers call themselves the to-Ussu.They maintain a complex of sacred sites which they associatewith the birthplace of all humanity and, by implication, theorigin of their Bugis ancestors as the original people. Ho wmany centuries ago Bugis speakers colonised Ussu, andwhether they originated this creation mythology or absorbed itfrom previous denizens , are vexatious questions that maynever be satisfactorily answered (footnote 68).

    . ula and the abutt ing coasta l lowlands profess to beinglim. Bugis communities have also sprung up right across- and outheast Asia. Their common ethnic identity, based on-hared language, a single religion, and observance of core:; alues, overrides minor linguistic differences and local. rural varia tion . onetheless there is also a clearly- rvable tendency for Bugis to identify by dialect to-WajoWajo person, to-Luwu etc.) , and to engage more readi lytb peakers of the same dialect. n South Sulawesi, when aguage is geographically circumscribed it equates to ethnicmiry, while far-flung languages promote ethnic identity atlevel o f the dialect as well as the language.

    YIany of these ethnographically observed features wereli hed by late pre-Islamic times, as reflected in the:;enous literature of the Makasars, who occupy the southouthwest fringes o f the South Sulawesi peninsula, andBugis. All local texts in the South Sulawesi peninsula andme m Luwu are written in Makasar or Bugis, depending onh language is locally dominant, implying long-term,graphically stable, core distributions for both languages.d. the spatial distribution of most Makasar and Bugisects corresponds to chiefdoms which feature in the pre

    -Iami literature. Further, in the fif teenth and sixteenthruries, people along the south and southwest coastal plain- me peninsula inhumed their deceased in an east-westtat ion, while people in the centre and north of theula, and at Malangke, cremated their dead. That is, attime of their conversion to Islam, the Makasars inhumedhile the Bugis cremated. In summary, the majority of thenin ula s inhabitants (whether Bugis or Makasar) have longnrified themselves a t the higher level as speakers o f thesam language, associated with shared religious (mortuary)ra rices, while at the local level they identify as members o f. ir chiefdom.Two social s trata prevail among traditional Bugis_Iaka ars, and some other South Sulawesi lowlanders with a

    -tmilar organisation: the red-blooded commoners, and the-tocrats who claim a proportion of white blood. Their whiteoe d supposedly derives from the tomanurung wh oulously appeared at various places across the South- wesi lowlands to institute a ruling lineage. The/lunmg are themselves stratified into the ancestors of the

    of the major chiefdoms, who constitute the royalty, and10IIlanurung ensconced within or slotted between thehiefdoms. Demonstrating the patrilineal or ambilinealem o f a prominent ruler from a renowned tomanurung isoccupation of the indigenous literature. Some noblescrnon trate their pedigree through descent from a royalonage, i.e. with reference to the main chiefdoms known:hroughout South Sulawesi. obles who can merely trace a

    cent relationship to a local tomanurung are more;: graphjcally circumscribed in their claims on privilege. Intb cases, members of the nobility rely on the sufferance ofdominant chiefdom, either through their relation to a royalurt. or as apical figures within local hierarchies that have-i -ed the fourteenth century and later expansion of the- chiefdoms.

    n the period leading up to the seventeenth century, as-trated in the genealogical records, royal mar riagesingly connected geographically distant polities. Whileay be ext reme to describe the South Sulawesi royalty asbecoming one extended family, it would certainly beurate to recognise royal coalitions, allied tlu ough familylationships, that committed ever larger numbers o f subjectsow ar ds c om mo n goals. IS This traditional sociopoli ticalorganization of an inter-married royalty whose members makee i ions of strategic significance, and local populationsd pendent on their ruler, has allowed social transformations to-weep rapidly across the landscape. This is best illustrated by

    5

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    4/14

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    5/14

    in the vicinity of Lake Matano compared to their occurrenceat sites below the 100 metre contour (Table 2; Fig. 1 . Theoldes t reliably identified wares date to the thirteenth andfourteenth centuries, as is the general pattern across SouthSulawesi. 27 Wotu (more precisely, its coastal satel lite ofTambu-Tambu) may have been Luwu s major coastal centreduring this period, as more thirteenth to fourteenth centuryimported ceramics were recorded here than in all otber Luwusites combined. The wealth o f ceramics at Tambu-Tambu,collected during a merely cursory survey of a disturbedcemetery of cremated burials, continued through to thesixteenth century. Wotu s trading importance stems from itsposition at the terminus of a traditional, major overland routethat crosses Sulawesi from the Gulf o f Tomini (due north ofLuwu). Its long-distance trade links are clear from the statusof the Wotu language as an isolate with vague affinities toLaiyolo (Selayar Island, 300 km due south of Luwu) and theButon languages in Southeast Sulawesi. Tampinna, thenearest site to Wotu, shows a similar maritime orientation. Byat least the fifteenth century Tampinna had been established asa Bajau sea gypsy settlement, and remained so until itsseventeenth century devastation at the hands of Wotu. 30However, Malangke, which lies within the continuousdistribution of the Bugis language, appears to have beenWotu s early counterpart. Tompe, Tampung Jawa andLindrunge, all in Malangke, are the only three Luwu sitesother than Tambu-Tambu to have clearly yielded thirteenthcentury imported ceramics.

    Fourteenth century ceramics occur far more frequently inLuwu than thirteenth century ceramics, in terms o f the numberof sites and the quantity at any site. Puang Ma tene and PuangBalubu (near Pinanto) , Ussu, Pinanto, the area o f Wotu spresent-day township, and three more Malangke sites join theranks of ceramic-bearing sites from the fourteenth centuryonwards. The pattern continues into the fifteenth and sixteenthcenturies with the recovery of Ming-period ceramics at sitesright across the study area, even extending to Lake Matano.This continuous increase in the importation of ceramicscorresponds to the historical indications (discussed below)

    e 2: Early historical sites in Luwu dated by imported ceramics 31Xs indicate major representations)

    The smel ting deposits at uha are dated to c. 1000-1300 atir base and 1700-1940 at the top. The smelt ing deposits at- oyu, sandwiched between a basal date of c. 1000-1300 andventeentb century Chinese sherd at the top, correspond toarne period (Table 1). This is also the general period ofdated grave goods frorn Pontanoa Bangka which includecotton mentioned above, 262 mainly Chinese glass beads. twelfth to fourteenth century antiquity, an eighteenthnrury Chinese coin, and an eighteenth century Chineseu er. Accordingly, Pontanoa Bangka would appear to haveorne the local cemetery when the smelters along Lakelatano s northern shore established their operations at uhad ukoyu during the second millennium Diron smelting evidently commenced at Matano at around

    : 00. Eight radiocarbon samples from the base of the smeltingits at Rahampu u and Pandai Besi have produced a suite.. tatistically indistinguishable dates whose means range\ een 1470 and 1640 (Table I). The earliest securely dateddewares recorded at Matano and Pa angkaburu, aftern ive survey and excavation, are a few fifteenth to[eenth century sherds (Table 2). The Pandai Besi ,ampu u and Lernogola excavations sampled a thickrrnation o f iron ore and slag, prills (tiny balls up to 5 mmmeter) of pi g iron, charcoal, earth furnace lining,nware tuyeres , and flaked chert str ike-a-lights. This

    :hropogenic formation extends nearly one kilometre alongforeshore, and as mucb as 100 metres up the slope from the. . Its sheer extent suggests that a group of experienced ironr ers, probably from the north shore o f Lake Matano,- ended on Matano to se t up their industry. Although the- t suitable ore lies some kilometres to the north, Matanoproximity to arable land along the lake s southern shore,direct access to the trade route leading to Ussu andkang (via Turunan Damar). These advantages would_ m to have allowed the industry to expand far beyond itslier development in the vicinity of uha. 26

    _ o dated principally through imported ceramicsmeenth century and earlier imported ceramics are sparse

    Centuries > 4 > 5 > 6 >X X X X

    X X XX X

    X X X

    SiteTambu-Tambu WotuWotu townshipTampinnaUssuTurunan DamarKatue/PoloeSalabu t nol Pa angkaburuSukoyu/Pontanoa BangkaPuang Ma tene/Puang Balubu (near Pinanto)PinantoSabbang Loang (later occupation)Benteng BaebuntaBenteng Tompottikka (Palopo)TompeTampung JawaLindrungePattimang TuaDadekoePincang PutePanasae (near Tompe)Utti BatueWalewalae/Mangge/Dato SulaimanLopa (near Mangge 1 Tanatede (near Pincang Pute)AratengLengkong Ulawengalangke Beccu

    XXX

    XX

    XXXXXX

    X XX X

    XX XX XX XX X

    X XX XX XX XX

    XX XX XX X

    XX X

    X

    171hXXXXXXXXXXXX

    X

    XX7

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    6/14

    that the peak of Luwu s effiorescence occurred between thefourteenth and sixteenth centuries.The boom period coincided with the development ofPinanto as a major iron-working centre. This 260 metre-long

    granodiorite ridge on the Baebunta River has short, steep sidesthat combine accessibility and ease of defense. Five of the 22test pits, from the north to the south o f the site, yielded ironslag, often associated with fragments of iron artifacts and, inone test pit, two iron prills. Iron ore was not recovered,suggesting that pig iron was brought to the site. Mostprobably, this was ore smelted in the vicinity of the BukitPorreo and Pangiwangen sources, and transported down theRongkong River to its junction with the Baebunta. Pinantowould have been linked to the small sites with fifteenth tosixteenth century ceramics at Sabbang Loang and BentengBaebunta (Table 2). The huge, looted site of Puang Ma tene,directly across the Baebunta from Pinanto, would have servedas the general cemetery for the area s population. Crematedhuman remains are also reported from two smaller, elevatedlocations which are the probable sites o f elite burial grounds the pinnacle on the Pinanto ridge and, upstream from Pinanto,Puang Balubu. The complex o f late pre-Islamic sites at theRongkong-Baebunta junction could be considered a collectingcentre, along the lines o f Bronson s model, but it may be morerelevant to emphasise the evident salience of iron. Pinanto ssecurity advantages, and its association with a private burialground, certainly suggest that the operators of its iron industryenjoyed great prestige.

    The oldest remembered Islamic grave at BentengBaebunta is locally associated with the chief of the Baebuntapeople. He was reportedly the cousin of the first sultan o fWare (Muhammad Wali Muzahir) and embraced Islam at thesame time 1605 .33 The archaeological evidence summarisedabove strongly suggests that this Baebunta-speaking polity

    f lour ished dur ing the f ifteenth and s ixt eenth centur ies .However , its developmen ts pale in comparison withMalangke, the pre-Islamjc capital of Ware , to be deta iled inthe next section.Major sociopolitical changes ensued after Ware embracedIslam. Malangke was virtually abandoned, while Palopo(Benteng Tompo tikka) regi ste red its f ir st evidence ofoccupation (Table 2). Local historical accounts record a

    succession dispute in the early seventeenth century betweentwo incumbents to the Ware throne, one based at Cilellelangin the south, and the other at Malangke. The outcome was thepermanent relocation o f the capital to Palopo, about halfwaybetween Cilellelang and Malangke. Of relevance here, a textcalled the Tributary and Domain List o fLuwuq, Bangkala andBinamu (TDLBB), which may be dated to c. 1600, lists thedependencies of Ware .34 Fully half o f them lie along thenarrow lowland strip that extends 70 south from Palopo.Hence the relocation of the capital should be understood as acompromise between the evident concentration of populationand agrarian potential in the lowlands south of Palopo, and thetrade resources in the highlands to the north. Moreover,Palopo is itself advantageously located at the terminus of amajor overland route that accesses resources such as gold andforest produce from the south-central highlands.3

    Although the archaeological evidence indicates that Luwucontinued to grow, in absolute terms, throughout the sixteenthcentury and perhaps beyond, it was overshadowed bydevelopments to the south. During the sixteenth andseventeenth centuries, the main agrarian Bugis chiefdomssuch as Wajo and especially Bone (Fig. 1 inset), and theexpanding empire of Macassar to the south, engaged inpeninsula-wide warfare. These battles increasingly left Luwuas victim or neutral bystander.36 Nonetheless, seventeenthcentury (and later) ceramics continued to be imported right

    ,Petta Pao i

    o[ kilometres

    5[

    Small habitation sites (extent not mapped)) Approximate location of uninspected,reported lootings Small, inspected looted areas\ Mapped looted areas- Mapped habitation sites/}. Monumental site

    ig 2: Pre Islamic sites recorded during the survey Malangke LuwlI.

    8

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    7/14

    Table 3. Estimated populations rounded off) for Malangke

    Site group 14 century 15 century 16 th century 17 centuryTampung JawaJPattimang Tua 2000 2000 2000 220Utti Batue cluster 270 5050 5050 0Peripheral sites2 430 2430 7500 660TOTAL (nearest hundred) 2700 9500 14500 900

    1 Lindrunge, Arateng, Mangge, Walewalae, Dato Sulaiman.2 Other burial grounds not mentioned above.

    Table 4. Estimated population size an d density of Southeast Asian cities

    City Population size Year Area (km2) Population density(persons/hectare)Thang-Iong , 1000 ODD 1640 22 455Thang-Iong 130 ODD 1688 22 59Mataram 8000005 1624 41 195Banten >100 0003 1672 5 >200Banten 220 0005 1673 5 440Banten 700 0003 1684 5 1400Ayutthaya 200 000

    31617-1686 15 147Aceh 45 500-52 000 1688 12 : : 41Macassar city 130 000 1650 12 108Patani 10-20 0003 1690 2 75

    1 Reid 1993:71-74 2 Bulbeck 19943 observer s direct estimate 4 based on 6. 5 persons per observer s estimated number of households5 based on quadrupling the estimated number of fighting men.

    MALANGKE EMPORIUM

    hectares (as the best estimate) dedicated to the iron industry.By the fifteenth century, Utti Batue had evidently emerged

    as Malangke s coastal trading centre. t is surrounded bylooted cemeteries o f predominantly fifteenth to sixteenthcentury antiquity (Table 2 . On e of these, Arateng, containeda boat-shaped coffin whose timber dates to between 1400 an d1635 (Table I). Although this coffm had c on ta in ed h um anashes , the three coffins at Mangge reportedly held inhumedburials oriented east-west. T he se w ou ld repr esent eit herMakasars or B aj au s ea gypsies. Th e Mangge cemetery alsoreportedly i nc luded a k c pi mandolin among its gravegoods (associated with a human skull). Th e s ou th ern area o fVtti Batue is said to h av e p rod u ce d a boulder e ng rav ed forplaying m c ng a B ug is g am e similar to chess. Utti Batuet hu s a pp ea rs to have been the centre o f a prosperous,c os mo po li ta n s oc ie ty w he re a t least some of the traditionalBugis arts a nd p as ti me s we re practised.39

    Th e tombs o f the first sul tan o f Ware , Muhammad WaliM uz ah ir, a nd his religious t ea ch er Dat o S u la im an lie at thenorth ofUtti Batue. However, the palace had evidently shiftedeastwards, to M a la n gk e B ec cu , by the early seventeenthcentury. Th e finest tradewares which date to c. 1600 occurhere, a lo n g wit h an early Islamic coffin exquisitely carved inthe B anten style ,40 an d a tomb elaborately decorated in a Hindu-Javanese style.1 Th e latter wa s the tomb of SultanNurussalam Petta M a la n gk e wh o ruled Ware at t he time tbecapital shifted to P al op o. Des pi te t he se signs of conspicuoussplendour, Malangke Beccu wa s clearly a m uc h s ma ll ersettlement than Utt i Batue, presumably reflecting the sbift o fsociopolitical power to southern Luwu.

    Tb e m ax im um p op ul at io n s ize o f the M al an gk e s it eco mp lex can be realistically estimated at ap pr ox im atel y10 00 0 residents. During inspection o f th e looted site s wen ot ed t hat the looter s holes, when still observable, weregenerally spaced three metres apart. This allows an estimate ofthe maximum number o f burials within any looted area, underthe assumption th at t hey w er e s pr ea d e ve nl y across eachcemetery. A series of conservative assumptions then allows an

    -- Luwu including Katue, apparently re-occupied by this. in association with the re-organisation o f Luwu into itsri ally well-attested form (Table 2).37

    OXIS p rog ram at Malangke included excavations at thelargest pre-Islamic habitat ion sites known in Luwu,an g Tu a (5.3 hectares) and Utti Batue 4. 0 hectares).

    personnel also m ap pe d n um er ou s looted areas3.riably associated with reports of fragments of cremated

    bone), an d inspected the unsold antiques and surface- from these sites. Th e spatial distribution of the sites ,

    burial o f Vtti B a tu e b en ea th half a metre o f postrb century alluvium, suggest that pre-Islamic Malangkeng a small ba y Figure 2).

    - the three sites in M al an gk e w it h c le ar traces o f anth century commencement date (Table 2), Tampung

    w as no t onl y Luwu s largest pre-Islamic cemetery, bu ti onI definite instance o f a pre-Islamic monument. Two

    m ou nd s, a pp rox i ma te ly 3 0 m et re s in d i am et er a nd sixhigh. had been ransacked by looters. Th es e m o rt ua ryument reportedly yielded cremated remains, an

    rmous variety o f gold, b ron ze a nd iron, and innumerablecs. Southeast of the m ou nd s, o n e or m o re b urie d b ri ck

    rures extended across a rectangular area approximatelymetres by 70 metres, and looters reaped antiques here too.- wou ld s ee m to be the sole pre-Islamic site in L uw u w it h

    ~ w o r There is every reason to associate Tam p un g J a waanese graveyard ) with a prosperous Javanese enclave,ially as Luwu is the o n ly B ug i s polity mentioned among

    fourteenth century vassals o f the J av an es e e m pi re o f. pahit .3-\t Pattimang Tua, across the r iver f ro m T am pu ng J aw a,

    avations recovered 34 iron implements and fragments,-h rds from i mp or te d c er am ic s s om e o f exceptionalI , which date back to the fourteenth century. Four o f the

    pit contained evidence o f iron working, indicating thatPalcnnlang Tua was a major industrial centre with possibly 1.69

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    8/14

    estimate of the populat ion represented .42 From thesecalculations, the Malangke population grew from an averageof 2 700 residents in the fourteenth century to an average of9 500 in the fifteenth century, and 4 500 in the sixteenthcentury, before plummeting to less than a thousand in theseventeenth (Table 3).

    These figures can be checked against the likely populationsize of the two surveyed Malangke settlements, based on acomparison with slightly later Southeast Asian cit ies.Seventeenth century European observers estimated both thepopulat ion size and area of seven Southeast Asian cit ies,hence these c it ie s populat ion densities can be derived(Table 4). Mos t of the resulting data points fall beneathFletcher s 43 empirical ly derived ceiling on the maximumpopulation density that can be sustained, for a givenpopulation size, within a single settlement (Fig. 3). In the caseof one of the exceptions, Thang-Iong, a much lowerpopulation estimate is also available. This lower figureprobably referred to the city proper, while the higher figureincluded the region as well as the city. A similar confusionprobably affects both of the graphed estimates for Banten,though this was undoubtedly a congested city, as all sourcesestimate a population in excess of 100 000. The estimate ofMataram s population is perhaps also too high, as it is basedon the number of fighting men (who would have been drawnfrom the surrounding region). All the other f igures appearreliable, especially in the case of Ayut thaya wi th fourinternally consistent est imates. The six or seven reliableestimates of the population densities in these Southeast Asiancities consistently fall between 40 and 200 persons perhectare, with the higher densities indicating congestedconditions (by world standards) for settlements of this size.

    Fletcher s study emphasises that small set tlements cansustain significantly higher population densities than largesettlements. Based on the interaction limit and the Southeast

    Asian examples already discussed, the densities could haveranged from an average of about person per hectare up toas many as 700 - 800 persons per hectare. In that case, for thevarious parts of Malangke, the 5.3 hectares in Pattimang Tuawould correspond to a population size between 530 and 3 700- 4 240 people, while the four hectares in Utti Batue wouldcorrespond to a figure between 400 and 2 800 - 3 200inhabitants. The Pattimang Tua figure agrees well with thepopulation estimate of 000 based on the Tampung Jawa andPattimang Tua cemeteries (Table 3). The estimated number ofinhabitants in Utti Batue is considerably less than theestimated population size derived from the surroundingcemeteries. However, the latter figure would include peoplewho were attached to Utti Batue but dwelled beyond itsboundaries. Similarly, up to half of Malangke s populationmay have resided in small settlements located away from themain centres of Pat timang Tua and Utti Batue (see the plussigns in Fig. 2). Hence the available data on settlement sizeconfirm the feasibility of the estimates in Table 3, and indicatethat the Malangke population size may have been in the -IS 000 range during Luwu s fifteenth to sixteenth centuryheyday.

    ot only was Malangke s population much larger thancould be the case for any other pre-Islamic site in Luwu, butalso Malangke s level of wealth must have been inordinatelyhigher. Looters report astounding quantities of ceramics, glassbeads, gold and other metals from the Malangke cemeteries.As noted above, Malangke evidently became a true entrepotwhich accommodated at least severa l language groups.Clearly, Malangke was the capital of Ware throughout thefifteenth and sixteenth centuries. Most probably, Ware wasbased at Malangke by AD 1365, when the Desawarnana theMajapahit text that mentions Luwu, was composed. Indeed,all of the recorded pre-Islamic rulers of Ware may haveoperated from Malangke. A 30-year average reign length

    A: Ayutthaya; B Banten; C: Aceh; D: Mataram; M: Macassar; P: Pantani; T: Thang-IongPersons per hectare(log scale)

    1000.Fletcher s bandofm ximuml _sustainable population density Bfor a given populationsizer _ :: _

    100 -

    10,000 100,000 I1 million millionNumber of people (log scale)

    Fig 3: Reported population size and density seventeenth cen LIry Southeast Asian cities compared to Fletcher s ceiling on sustainable populationdensity.

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    9/14

    ELITE MILIT RY O TROLThere is abundant archaeologica l evidence pomtmg to theimportance of iron weaponry in late pre-Islamic Luwu. Directevidence includes fragments of knives, machetes and asuspected kris recovered from Pat timang Tua, Utti Batue,Pinanto and Rahampu'u. Knives, swords and daggers featureamong the goods reportedly looted from Puang Ma' tene,Puang Balubu, Tampung Jawa, Pa ttimang Tua, Mangge,A ra te ng , D ade koe , P in ca ng Pute, Lengkong Ulaweng,Tampinna, Kubur an Uss u, Weilawi Malangke), andKawasule (upriver from Salabu). I was shown a two-metrelong cannon dug up at Wotu's pre-Is lamic cemetery, and acannonball from Mangge, both o f which imply the presence ofartillery in pre-Islamic LUWU. 7 The emphasis on weapons inthe historical accounts of Luwu iron (discussed earlier) isstrongly corroborated by the archaeological evidence.

    Security in Luwu s hinterland may have been a constantheadache. Late nineteenth century European observers paint aterrifying picture of the Mori and related linguistic groups inthe uplands north of Wotu. Their specific variant of rankedsocial organization involved constant raiding and, especially,headhunting to elevate the s ta tu s of any community s(temporary) leadership.58 The Rongkong and Matano ironworking communities may have depended for their viabilityon being major weapon producers. The Rongkong inhabitantsindeed enjoy the reputation of being among Luwu s fiercestwarriors. 59 The location of the main Baebunta settlement ofPinanto on an elevated ridge suggests concern with attacksfrom the mountains even where sites border the coastal plain.The hinterland dependencies o f Ware' (as depicted in theTDLBB) may have been restricted to Rongkong and Matanosimply because secure lines of communication could bemaintained in the highlands, and were only worth maintaining,with communities producing abundant weaponry.

    Communities along Luwu's coastal plain would have beenat some risk from highland assaults, and prone to squabblingwith each other, but security here would have beeneconomically viable. The Ware' troops, based in Malangke(where so many weapons have been looted), may have

    Lake Matano is the suspected source of the ore that givespamor luwu which is distinguished by silvery lines onpattern-welded kris daggers after they had been darkened withchemical etchants. Pamor luwu appears to have been used inJava to produce prest igious krisses since Majapahit times.54The fifteenth to sixteenth century establishment of a majoriron-smelting operation at Matano coincides with Malangke'sheyday.

    If the high stakes in Luwu s economy revolved on iron, asappears likely, how could Ware' have employed economicincent ives to concent ra te the iron industry in Malangke?Conceivably, rice may have been widely desired in Luwu forfeast ing and as an eli te foodstuff, and Malangke could havemonopolised the rice trade from the south. Nonetheless, exoticprestige goods, best represented archaeologically by ceramics,were probably more important. Fifteenth to sixteenth centurywares occur widely in Luwu sites, but large numbers havebeen documented only at Malangke, Baebunta and Wotu. Asnoted above, all relevant evidence indicates a closerelationship between Baebunta and Ware'. And Wotu probablyoperated as a coastal collection centre for produce, includingMatano iron (via Ussu), to be shipped on to Malangke. 6 Theshared practice of cremations at Baebunta, Malangke andWotu further suggests elite linkages between these threetrading nodes. Elite control over the distribution of prestigegoods is invoked in many models (including Earle 's) on thefonnation of complex polities, but this raises the question ofhow Ware' maintained that control.

    O TROL OVER THE LUWU_-OMYtraSt to the cases s tudied by Earle, food product ioneem to have offered little scope for manipula tivetrategies in Luwu. North of Palopo, the Luwu

    ..:.;. .;;.rll S are a major destina tion for both spontaneous and:lIllent-sponsored transmigration, which implies that its_ _ tural potential has been his tor ically under-exploited.thrives regardless o f human inducement across the vast

    ~ - Z = l D \ : reaches. Only its processing would have involved amvesrrnent of labour. In general Luwu stands out as aor arboriculture. Although parts ofLuwu are renownedniggardly soils on their ultrabasic fonnations, thishave affec ted only a few OXIS sites (Salabu and thean g Lake Matano's northern shore). The bones oflarge

    a ~ u c mammals (water-buffalo or cattle, and pigs) aren wherever preservation conditions allow (e.g. Salabu,L..- t .uua. Utti Batue), suggesting abundant pasture and brush- a lightly inhabited land.

    produce is an important considera tion and maytituted the bulk of Luwu's export trade. Dammarnan cane, ebony and gharu wood figure as major

    re ources. On e variety of the abundant bakau_ \ e produces a fine red timber, which may explain_ Tampinna repor tedly produced wooden sheaths for~ - r ' ~ r r 1 n t s of Matano iron. Wotu may have owed much of- - r trading prominence to vegetable products. However,r ubiqui ty of Luwu 's forest produce would havea prevent elite control over its distribution, even if ittituted a major sector of Malangke s non-elite

    _ igbtly sourced iron reserves in Rongkong (Bukitd Pangiwangen) and Matano would have been more. E C : c : ; ~ l e to chiefly interference. Of all the toponyms in Luwu

    _ the Tributary and Domain List of Luwuq, Bangkalau (TDLBB), only Rongkong and Matano lie well the metre contour.49 The major product of

    'hich in effect lies at the head of an enonnous deltao . . , ~ ; e d by Malangke, was surely forged Rongkong iron. Then industry at Pattimang Tua presumably utilised ironI rano as well as Rongkong. The to Ussu Bugis are

    .ES::rii:1 ed a mythological significance which almost certainly_- a the importance of the Ussu-Cerekang area as Luwu s,..---,... or Matano iron. _ 'ong iron ore was reportedly used only for making

    ...,...,.. n ,.-. Those of Porreo' iron supposedly poison the victimlh even if the blade penetrates the skin by a few

    ce:::::II::letres. The Pangiwangen iron is also renowned as- ' ' --grade ore. According to one story in Rongkong, the. and Pangiwangen ores were blended to produce all the~ : O ; X l n 5 - borne by important Luwu male aristocrats.52 Writing

    . an Braam Morris asserted that the most prized iron' I : : : O : ~ - in the entire Netherlands India were made in

    ' and Wotu. 53 From the archaeological evidence,iron would have merited a similar level ofduring the fifteenth and sixteenth centuries.

    ommence the inaugural reign of Tampabalusu to- a date which agrees remarkably well with thelogical evidence on Malangke's early expansion.angke is an ideal location for sago,46 and marine

    ,::- ,.n:I:S- uch as fish would have thrived in the mangrove- and adjacent bay. These resources may well have giveneenth to fourteenth century settlers a head star t, but- ardly account for the spectacular growth o f pre-Islamic

    , ~ ~ ~ _ ; r > c : . Fo r this, we employ Earle's analysis of the chieflylation o f the three main sources o f power.

    11

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    10/14

    col lec ted protect ion payments to secure Luwu s coastallyoriented dependencies or, during calmer interludes, to leavethem in peace.60 Control over the distribution of the weaponsmade of Rongkong and Matano iron, and a large standingforce in Malangke (supplemented by warriors from thesmaller settlements), would both have been necessary toensure that Malangke retained its military control. avaldominance keeping the seaways safe, and even proactiveencouragement to use the Malangke facil it ies, are stronglyimplied by Malangke's commercial success.

    Recent work suggests that the founding of Malangke mayhave been an extension of the early naval prominence ofWare' . The list of Ware' rulers starts with three apocryphalfigures who are also associated with Cina, an ancient Bugischiefdom well to the south of Luwu (see Fig. I inset). TheCina genealogy claims that Simpurusia (the first ruler ofWare') descended from heaven at Lompo.6 According to localaccounts collected by Tan Caldwell, Lompo lies in Sengkang,the capital of modem-day Wajo. Ware' could have colonisedMalangke as part of a naval expansion that involved controlover the mighty Cenrana River (which flows from Sengkangto the coast) and the coastline as far north as Malangke.

    Caldwell's hypothesis neatly explains a puzzle in Bugispre-Islamic history. At approximately AD 1500, Ware' wasinvolved in a series of confl icts with Wajo, Sidenreng andBone, three Bugis agrarian chiefdoms located south of Luwu(Fig. I inset). In the mid sixteenth century, Bone forced Ware'to abandon the mouth of the Cenrana. The accounts indicatethat Ware' was gradually forced to surrender land that it hadpreviously controlled. Previous scenarios have alwaysinterpreted these records as the repulsion of an unrecorded butancient expansion by Luwu into the southwest peninsula62 Aless presumptuous interpretation is afforded by the evidencethat Ware' had originated on the Cenrana. Th e battlesdescribed would now be seen as the staged transfer of theancestral domains of Ware' to the agrarian chiefdoms whichflourished along the Cenrana Valley af ter c 1400. Theexpansion by Ware' into Luwu may even have stretched itsmilitary resources, leading to the loss of the Cenrananotwithstanding the net territorial gains.ELITE ONTROL OVER I EOLOGYFour intertwined e lements may be detected in the eliteideology of Ware '. These are a rich array of Indic overtones,the standard Bugis emphasis on the white-blooded pedigree ofits rulers, Luwu s primordial status in the La Galigo epic, andthe organisation of Luwu into directly ruled lands (domains)and tributaries in the TDLBB.

    Luwu's connections with late Hindu-Buddhist Java areimplied by its citat ion as a Majapahit vassal, and the Sanskri tnames of two of its rulers.6 This raises the question of thedegree to which Ware' adopted Hindu-Javanese culturalforn1s. Caldwell's reconstruction of early Ware' as a marit imepower controll ing the Cenrana Valley and southern Luwusuggests that Ware' was more open to long-distance contactsthan the other Bugis chiefdoms. Ware' may have derivedconsiderable prestige from its links with Java, while theJavanese at Tampung Jawa would have contributed technicaland organisational skills to Malangke's operations. Ware' isthus a likely conduit for the smattering of Indic elementsfound in traditional Bugis cul ture, such as the name of thetransvestite priests bissu , the days of the months, and certainritual and cosmological aspects of rice cultivation. However,Indic high culture, such as the Ramayana and other literaryepics, monumental inscriptions, and any concepts of priest andwarrior castes, would appear absent from South Sulawesi.64Any Hindu-Javanese ideological influence on Ware' wouldseem to have been superficial. 2

    So much is clear from origin stories about Ware' . On e setof legends, the attoriolonna Simpurusia, stresses relationswith Majapahit. also combines numerous Sanskrit emblemsand names with possib le evocations of elements from theMalay version of the Ramayana. Bu t the point of theselegends is to account for the regalia of Ware' , reflect ing astandard Bugis literary concern with royal insignia. Twogenealogical versions of the apocryphal founder of Ware',Simpurusia, emphasise his association with settlements alongthe Cenrana.6 Though his name may well be a corrupt ion ofthe Sanskrit for 'lion man the important point is hisrepresentation as a 'descended one or tomanurung, analogousto the legendary founders of the other Bugis chiefdoms. Hencehis descendants, the rulers of Ware' , locked into usual Bugiselite ideology regarding the basis of their legitimacy to reign.

    However, the late pre-Islamic history of Ware' was, ingross terms, the loss of the Cenrana as Ware' consolidated itshold over Luwu. This trajectory poses problems of legitimacy,in terms of ancestral occupation, both for the chiefdoms whichabsorbed the ancient domain of Ware', and for Ware' in itsnew holdings. I suggest that the La Galigo mythology mayhave been manipula ted as an ideologica l device to resolvethose problems.

    The La Galigo, one of the world 's largest bodies ofliterature, was evidently not transcribed before the eighteenthand nineteenth centuries . However, these t ranscrip tionsemploy a form of Bu gis that dates back to at least the sixteenthcentury. By the late seventeenth century the concept of a longgone L a Galigo age had developed.66 The events in the LaGaligo start with the descent to earth of Bataru Guru, son ofthe ruler of the UpperWorld, to marry the daughter of the lordof the Lower World. Th e fourth generation o f theirdescendants includes the main protagonist of the La Galigo,Sawerigading. After adventitiously falling in love with hissister, Sawerigading hears of a princess of Cina, We Cudai, ofidentical appearance. He ventures from Luwu on an epicvoyage to abduct We Cudai, and most of the subsequent actiontakes place in Cina. After the birth of the sixth generation, allthe descendants leave earth, except for one couple who stayand rule over Luwu' The story thus seemingly originatesWare' in Luwu, and transports Ware' into the Cenrana and outof it again during a bygone time popula ted by semi-div ineheroes.

    The to-Ussu treat the La Gal igo as the account of howhumanity originated in Cerekang. Their sacred sites locate themajor events of the La Galigo within Ussu and Cerekang.Wotu speakers, however, challenge to-Ussu ownership bylocating the early events of the La Gal igo in Wotu. 68 The LaGaligo creat ion myth presumably originated somewherebetween Wotu and Ussu, especially as many South SulawesiBajau claim Ussu as their place of origins, in a group of storiesthat contain numerous La Galigo elements.69 Whoever mighthave first formulated the creation mythology, its attractions tothe Ware' court are obvious. could be construed asestablishing the primordial, divine authori ty of the Ware 'rulers in Luwu, prior to their appearance in the Cenrana, andit emphasised the special status of the Ussu area as Luwu smain outlet for Matano iron.Finally, the Tributary and Domain List of Luwuq, Bangkalaand Binamu (TDLBB) reflects pragmatic aspects in the eli teideology of Ware'. Bangkala and Binamu were two Makasarspeaking chiefdoms on the very southwest tip of Sulawesi, anideal staging post for trade with the western archipelago. TheTDLBB mentions them probably as a reflection of theircommercial relat ionship with Ware . In Bangkala andBinamu the tributaries an d domains fall into sensiblegeographical groupings which would appear to ref lec t thehistorical expansion of these polities.'o However, the concepts

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    11/14

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    12/14

    2122

    23242526272829303132

    3334353637

    38

    39404142

    14

    Pel ra s 1996; C al dw el l a nd D ru ce 1998; Koolhof 1999;Darmawan et a1 1999.Willems 1938) excavated II massive earthen jars fromthe same locality in Sabbang Loang referred to here, andsubsequent infrastructural developments have disturbedmany more. Willems also mentioned Chinese sherds ,but he presumably obtained these from a separatelocation in the site where i mpo rt ed , f if tee nt h toseventeenth century antiquity ceramics may be found assurface finds Bulbeck and Prasetyo 1999, 2000).Bulbeck and Prasetyo 1999; Bulbeck 2000.Bulbeck 2000; see Bulbeck and Prasetyo 2000.Bulbeck 2000.Bulbeck and Prasetyo 1999; Bulbeck 2000.Bulbeck 1992, 1996-7.Pelras 1996:60.Grimes and Grimes 1987; Sirk 1988.Caldwell and Druce 1998:23; Bulbeck and Prasetyo1999.Bulbeck and Prasetyo 1999.B ul be ck 2 00 0; B ul be ck a nd Pra se ty o 1999. As furth erevidence o f t he h ig h sta tu s o f ironworking, one o f theimported jars looted at Puang Balubu reportedly held thetools of an iron smith.Unpublished field notes.Caldwell and Druce 1998.Bulbeck and Prasetyo 1999; see also Caldwell 1988.Andaya 1981; Pelras 1996.Consideration of the Islamic period lies beyond thescope of t hi s paper. The i nt ereste d rea de r may c on su ltMappasunda and Hafid 1992/3), Irfan 1993) andSchrauwers 1997).Robson 1995; Bul bec k and P rasetyo 1999. Strictlyspeaking, the meaning of Tampung Jawa in Bugis isforeigner s graveyard, but this is because the Javaneserepresent the archetypal foreigner to the Bugis.Bulbeck and Prasetyo 1999 2000.See Ambary l 1993: Plates 153-162.Van Lijf 1953. The tomb was destroyed during the1950s Kahar Muzakar rebellion.Assume that, at Malangke, the number of looted burialsbroadly equals the number of buried adults. Assume alsothat a Malangke inhabitant, having reached adulthood,had an average life expectancy of 40 to 45 ye ar s ol d cf.Acsadi and Nemeskeri 1970:213), i.e., this person livedas an a dul t in Ma la ngk e for a bo ut 25 years. He nc e f ou rl oo te d b uria ls w ou ld o n a ve ra ge rep re se nt o ne a du lt inthe population for a period of 100 years. T he se f ou rb uria ls w ou ld furth er c orre sp on d t o an a ve ra ge of twoindividuals 1 adult I child) over the century, based onthe conservative assumption of equal numbers of adultsand children in the population at any time cf. Siven1991: 115). That is, if the c em et er y had been used for acentury, the number of looted bur ials di vid ed by t wow ou ld rou gh ly rep re se nt t he a ve ra ge p op ul at io n u si ngthe c em et er y o ve r the century. T he s ur ve ye d s ize of alooted burial ground allows us to estimate the number ofb uria ls, spa ce d t hree met re s a pa rt , w it hi n t he c emet ery call thi s B). D iv id in g this fig ure b y t wo B /2 ) p ro du ce sthe average population the cemetery represents if it hadb ee n used for a ce ntury. H ow ev er, most c emet erie s hadbeen used for mor e t han a c entu ry Table 2). H en ce thevariable B/2 is divided by the number o f centuriesd urin g w hi ch t he c emet ery w as in use, to e st im at e the

    43

    44

    45464748

    49505152535455

    56

    57585960

    616263

    a ve ra ge p op ul at io n represented by the cemeteryt hrou gh ou t its p erio d o f use Bulbe ck and Prasetyo1999).F let che r 1995. I th an k R ola nd F le tc he r for his advi ce,fol lo we d in t he text, o n h ow to i nt erpret t he somet imesconflicting estimates of the p opu la ti on s iz e o fseventeenth century Southeast Asian cities.My use of Reid s figures Table 4) excludes allp op ul at io n e st imat es b ased o n t he n umbe r of fightingmen, except in the case of Ma ta ra m whe re no o th erestimate is available), and Banten where I haveemployed the median of three figures ranging between125 000 and 800 000). N ot e that the e sti mat e ba se d ont ro op n umbe rs w ou ld a pp ea r to be rel ia bl e in t he c ase ofAyutthaya 240 000 compared to several direct estimatesof a population of 200 000). T he surprisingly highpopulation density in B an te n c ou ld n ot be a sc ri be d t o a nunderestimate o f the city s extent, as the archaeologicalsurvey and excavations of Banten would imply a slightlys ma ll er a re a o f only some 4.5 square km NayatiI994:Map 4). Reid 1993:73) also estimated a maximumurban congestion of a pp ro xi ma te ly 20 0 per so ns p erhectare in seventeenth century Southeast Asia.Caldwell 1998:39.Osazawa 1986.Caldwell 1993; Caldwell 1995:411; Caldwell and Druce1998:22-23; Bulbeck and Prasetyo 1999.To some degree, the iron industry figures so prominentlyin Luwu s reconstructible archaeology because of thep oor pr es er vati on p ros pe ct s for the r eg io n s or gan icexports. Analysis currently underway on the phytolithsfrom various sites should provide an insight into thed eg re e to w hi ch Luwu s forest produce complementedthe trade in iron.Bulbeck and Prasetyo 1999:Fig. IPel ra s 1 99 6:5 9. A cc ordi ng to Pel ra s, M at an o iron w asknown in L uw u as the iron of the Ussu people.Caldwell and Druce 1998:16.Bulbeck and Prasetyo 1997.Mappasanda and Hafid 1992/3:15.Bronson 1987; Maisey 1988; Caldwell 1988:183.Matano however presented an administrative problemfor War e , as it wa s remote and its iron could be just ase asil y e xp orte d n orth wa rd s, rat he r t ha n sou th t hrou ghLuwu. P elr as 19 66 :2 49 ) links th e de cl in e in Luwu shegemony after c 1500 to Luwu s l oo se ni ng g ri p onMatano iron.Wotu a pp ea rs to h av e e nj oy ed c on si de ra bl e a ut on omywithin Luwu, hence its exclusion from the dependencieso f War e listed in the TDL BB Caldwell and Dr uce1998). Sirk 1 98 8) e ve n sug ge st s t ha t Wotu w as one ofthe languages of the Luwu court.Bulbeck 1996-7:1041; Bulbeck and Prasetyo 1999.Schrauwers 1997.Mappasanda and Hafid 1992/3:51.Tampi=a s devestation by Wotu in the late seventeenthcentury ma y well reflect Luwu s diminishedp ea ce ke ep in g c apabi lit ies af ter the capital s hi fte d toPalopo.Caldwell 1988:92, 1998.Caldwell 1988: 190-194; Pelras 1996.Uniquely among the Bugis chiefdoms, two pre-Islamicrulers of Ware Batara Guru and Dewaraja, boreSanskrit titles Caldwell 1998).

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    13/14

    July 1994, Perth.BULBECK, F. D. 1996. T he politics of marriage and themarriage o f polities in G ow a, Sou th Sul aw esi, d ur in g t he

    16 h an d d h centuries , in J.1. Fox and C. Satther eds)Origins Ancestry nd Alliance: Explorations inAustronesian Ethnography Australian ationalUniversity, Canberra, pp. 280-315.

    BULBECK, D. 1996-7. T he Bronze-Iron Ag e of SouthSulawesi, Indonesia: mortuary traditions, metallurgy andtrade , in F.D. Bulbeck and N. Barnard eds) AncientChinese nd Southeast Asian Bronze ge CulturesSouthern Materials Center Inc., Taipei, pp. 1007-1076.

    BULBECK, D. 2000. T he I ron G at es of Indonesia. Twomillennia o f iron metallurgy in L uw u, Sou th Sulawesi ,p ap er p re se nt ed at the W or ks ho p on t he E ar ly H is to ry o fSouth Sulawesi, 18-19 August 2000, Canberra.

    BULBECK, F. D. and B. PRASETYO. 1997. Inspection o fLuwu historical sites 10-23 November 1997. R ep or t t oPusat Penelitian Arkeologi Nasional.

    BULBECK, F. D. and B. PRASETYO. 1998. Survey of preI sl am ic h is to ri ca l si te s in L uwu , Sout h Sulawesi ,Walennae 1:29-42.BULBECK, F. D. and B. PRASETYO. 1999. Th e Origins o fC omp le x Soc ie ty in South Sul awesi OXIS): tenta tive[ ma l r ep or t to Lembaga Ilmu Pengetahuan Indonesia.

    BULBECK, F. D. and B. PRASETYO. 2000. T wo millenniao f socio-cultural development in L uw u, Sou th Sul aw esi,Indonesia , World Archaeology 32:121 137

    CALDWELL, 1. A. 19 88. Sou th Sul aw esi A .D . 1 30 0- 16 00 :Ten Bugis texts, Ph.D. thesis, A u str alian N atio nalUniversity, Canberra.

    CALDWELL, 1. 1993. Untitled, Baruga 9:6 8.CALDWELL, 1. 1994. A r ep or t on f ie ld wo rk in L uw u w it h

    Bahru Kallupa and I wa n S uma nt ri in August 1994 ,Baruga 10:9 10

    CALDWELL, 1. 1995. Power, sta te a nd soc ie ty among thepre-Islamic Bugis , Bijdragen tot de Taal Land enVolkenkunde 151 :396-421.

    CALDWELL, 1. A. 1998. T he chronology o f t he k ing list o fLuwu to AD 1611 , in K. Rob in so n a nd Mukhl is Paeni eds) Living Through Histories: Culture HistOly andSocial Life in South Sulawesi Australian ationalUniversity, Canberra, pp. 29-42.

    CALDWELL, 1. and S. DRUCE. 1998. Th e tributary anddomain lists o f Luwuq, Binamu a nd Ban gk al a. Rep or t tothe South-East Asian Panel o f the British Academy.

    CALDWELL, 1. and M. LILLIE. In press. Manuel Pinto sinland sea: using palaeoenvironmental techniques toa ss es s historical e vi de nc e fr om S ou th Sulawesi , forModern Quaternary Research in Southeast Asia 7

    CALDWELL, 1. and W A. BOUGAS In p re p. T he rise ofBinamu and Bangkala in Jeneponto, South Sulawesi 1300-1500 A.D.).

    CHA BO T, H. T. 1967. Bontoramba: a village of G oa , Sou thSulawesi , in Koentjaraningrat ed.) Villages in IndonesiaCornell Press, Ithaca, Ne w York, pp. 189-209.

    DARMAWA M. R., SARITA P., GUFRA D. D. and D. D.A DAYA 1. 1999. O XI S: The Origins o f ComplexSoc ie ty in South Sulawesi A Social AnthropologicalPerspective), Australian National University TheA ustr al ia n Resea rc h Cou nc il ) and Universitas NegeriMakassar, Makassar.

    EARLE, T. 1997. How Chiefs Come to Power: The PoliticalEconomy in Prehistory S ta nf or d U ni ve rs it y Press,Stanford.

    eLl 1995:403.. 1 98 8: 42 -4 7, 6 2, 92. One of the settlements,

    pa ng eng, p os si bl y reappears as a dependency of. in the TDLBB, w hi ch w ou ld b e a na ch ro ni st ic a soilier et tle men ts a long the Cenrana ar e cited

    ~ e l l an d Druce 1998). Ware court literature ma yr tained the memory of the chiefdom s ancient. g- even if t he y h ad b ee n sun ke n i nt o t he l eg en da ryknight 1993:26-28; Koolhof 1999.

    ~ k n i g h t 1993:26; Pelras: 1996; Koolhof 1999:371.w el l a nd Druce 1998; Darrnawan et al. 1999. There

    iderable anecdotal evidence to suggest some levelcoin idence between the pre-Islamic historical sites ini ll a nd C er eka ng a nd the area s sacred sites Bulbeck

    Pra etyo 1998, 1999). OXIS was barred fromrki ng in the s acr ed sites, w hi ch prevent s a re li able- h aeo log ical p ers p ectiv e on the s et tl em en t a nd_ nomic history of t he a rea . U nt il t ho se sac re d sit es a re

    d to a rc ha eo lo gi ca l w or k, t he a nt iq ui ty o f Bugisment h er e, a nd t he ir r el at io n to t he t ra de in Matanowill remain mysterious.> ner 1998.~ e l l and Druce 1998; Caldwell and Bougas in prep.

    k an d Prasetyo 1999:Fig. 1.k 1993.

    . l ac kn ig ht 19 93; Cal dw el l 1 995 ; Pel ra s 19 96.Earle 1997:209-210.

    :21

    Se arc hi ng for g ood fortune: Theo f a Bugis shore community at L ak e L in du ,

    I u la we si , P h. D. t he si s, A us tr al ia n ational_ ity. C an be rr a._ G. and 1. NEMESKERI. 1970. History ofHumann and Mortality Akademia Kiad6, Budapest.

    lLLAH, M. 2000. T he Islamic antiquities o f.:=: A contribution to history of the Bugis community. IValennae 4: 111-135. M. S. TAKASHI and M. T. . H. WIBISONO.3anten Pelahuhan Keramik Jepang: Situs Kotaall islam di indonesia Pusat Penelitian Arkeologi

    -i::;;:;OmI. Jakarta.L. 1981. The Heritage o fArung Palakka Martinus- The Hague.. B. 1977. Exchange at the upstream and>treaIT1 ends: notes towards a functional model of the

    - ta te in S ou th ea st Asia , in H ut te re r, K .L . e d.)- ,,..,.- .w,,;c Exchange and Social interaction in Southeast

    _ n iv er si ty o f Michigan, Ann Arbor, pp. 39-52._ . -. B. 1987. Terrestri al and met eoric iron in the

    - >- ia n kr is , Journal o f Historical Metallurgy_ . - . . u ~ . ~ F.D. 1992. A tale of two kingdoms: The historical

    ~ : ; : .....t ~ o y o f Gowa and Tallok, S ou th S ul aw es i,- ia . Ph.D. t he si s, A ustr al ia n N at io na l U ni ve rsit y,

    . D. 1993. N ew perspectives on e ar ly S ou th-i hi tory , Baruga 9: I 0-18.

    L D. 1994. E co lo gi ca l p ar am et er s o f settlementan d hierarchy in t he p re -c ol oni al Macassar

    . l i T - ~ < i ~ ~ . paper presented at th e Asi an St ud ie s~ ; o c i a r i o n o f A ustr al asia Bie nn ia l Con fe re nc e, 1 3- 16

    Jo..::LOI.l.... vGRAPHY

    s

    f

    15

  • 8/13/2019 Economy Militery Pre Islamic Luwu-18_04_Bulbeck

    14/14

    F LE TC HE R, R. 1995. The Limits Settlement Growth,Cambridge University Press, Cambridge.GRIMES, C. E. and B. D. GRIMES. J987. Languages South Sulawesi, Australian National University, Canberra.IR FA N M., M. 1993. S tr uk tu r kot a P al op o a ba d X VI I- XI XMasehi: studi arkeologi tentang pemahaman eksperiensialdan cita- pikir an, Doctorandus s ubthes is , H as anuddinUniversity, Ujung Pandang.KOOLHOF, S 1999. T he L a Galigo a Bugis encyclopediaand its growth Bijdragen tot de Taal Land- enVolkenkunde 155:362-387.KOOLHOF, S 2000. R ei ze n in een e inde loo s v er ha al , inH.A. Poeze ed.) Oosterse omzwervingen: Klassieketeksten over Indonesiij uit Oost en West, KITLVUitgeverij, Leiden, pp. 15-49.LIEBN ER, H. 1998. Four oral versions of a st ory a bo ut theorigin of the B aj o pe ople of southern Selayar in K

    Robinson and Mukhlis Paeni e ds) Living ThroughHistories: Culture, HistOly and Social Life in SouthSulawesi, A us tr al ia n a ti onal Uni ve rsi ty, C an be rr a, pp.107-133.

    MACKE ZIE, C. 1994. Origins of r es ist anc e: t heconstruction and continuity of identity in Tana Towa B ul uk um ba , S ou th S ul awe si , I nd on es ia ), M.A. th esis,Australian National University, CanbelTa.MACKNIGHT, C. C 1983. T he rise of agriculture in South

    Sulawesi before 1600 Review Indonesian andMalaysian Affairs 17:92-1 J6MACKNIGHT, C C 1993. Th e early history of SouthSulawesi: some r ec en t a dv an ce s, M on as h U ni ve rs it yCentre of Sou th ea st Asia n Studies worki ng pa pe r 81,Clayton, Victoria.MAISEY, A. 1988. Kris and meteorites Journal Historical Metallurgy 22:58-59.MAPPASUNDA, H and M Y HAFID. 1992/3. KerajaanLuwu Menurut Catatan D.F Van Braam Morris , BalaiKajian Sejarah dan Nilai Tradisional, DepartemenPendidikan dan Kebudayaan, Makassar.NAYATI, W 1994. The archaeology of trading sites in theIndonesian archipelago in the s ixteenth and s eventeenthcenturies: possibilities and limitations of the evidence,M.A. thesis, Australian ational University, Canberra.OSAZAWA, K 1986. Sago production in kabupaten Luwu,South Sulawesi - a trial for upgrading the economiccapability of a traditional s ago- pr oducing s ociety , in K.Tanaka et al. eds) Environment, Landuse and Society inWallacea, K yoto U nivers ity Center f or S outheast A sian

    Studies, Kyoto, pp. 51-63.P EL RA S, C. 1985. R el ig io n, t ra di ti on an d t he d yn am ic s o fIslamization in South Sulawesi , ArchipeI29:107-135.PELRAS, C. 1996. The Bugis, Blackwell, Oxford.REID, A 1993. Southeast Asia in the Age Commerce I4501680. Volume Two: Expansion and Crisis, Yale UniversityPress, Ne w Haven.ROBSON, S translator) 995 Desawamana Nagarakrtagamaby Mpu Prapaiica, KlTLV Press, Leiden.SCHRAUWERS, A 1997. Houses, hierarchy, headhuntingand exchange: rethinking political relations in theSoutheast Asian realm of Luwu Bijdragen tot de TaalLand- en Volkenkunde 153:356-380.S IRK, U. 1988. Research needs ; the Wotu language , Baruga

    2: 10-12.SIVEN C. H. 1991. O n estimating m or ta li ti es f ro mosteological data International Journal Anthropology

    6:97-110.

    TUIVER, M. and P REIMER. 1993. Extended Cdatabase and revised CALIB radiocarbon dating programRadiocarbon 35:215-230.VAN LIJF, M. 1953. K or te me de de li ng : H et graf van deL uw u s vorst Ma ti nroe ri Malangke Bijdragen tot de

    Taal Land- en Volkenkunde 109:379-380.WHITTE , A J., M. MUSTAFA and G. S HENDERSON.

    1987. The Ecology Sulawesi, Gadjah Mada UniversityPress, Yogyakarta.WILLEMS W A 1938. Preliminary report on theexcavation of an urn bur ial ground at Sa bang near Palopo Central Celebes) , in F.N. Chasen and M.W.F. Tweedie eds) Proceedings the Third Congress Prehistorians

    the Far East, Government of the Straits Settlements,Singapore, pp. 207-208.


Recommended