Shailendra origins

This article was downloaded by: [Institutional Subscription Access]On: 01 September 2011, At: 17:12Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Indonesia and the Malay WorldPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/cimw20

Why the Śailendras were not aJavanese dynastyRoy E. Jordaan

Available online: 22 Jan 2007

To cite this article: Roy E. Jordaan (2006): Why the Śailendras were not a Javanese dynasty,Indonesia and the Malay World, 34:98, 3-22

To link to this article: http://dx.doi.org/10.1080/13639810600650711

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching and private study purposes. Anysubstantial or systematic reproduction, re-distribution, re-selling, loan, sub-licensing,systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representationthat the contents will be complete or accurate or up to date. The accuracy of anyinstructions, formulae and drug doses should be independently verified with primarysources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

http://www.tandfonline.com/loi/cimw20

http://dx.doi.org/10.1080/13639810600650711

http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/page/terms-and-conditions

Roy E. Jordaan

WHY THE SAILENDRAS WERE NOT A

JAVANESE DYNASTY

Archaeologists and art historians generally agree that most of the 8th- to 9th-centuryBuddhist temples in central Java were constructed by the rulers who claimed to belongto the Sailendra dynasty. But the unsolved question preying on the minds of scholarsfor many decades now is the origins of this dynasty. Drawing on a variety of argumentsthis study contends that the present popularity of the Sailendra dynasty’s Javanese originis ill-founded, and urges a resumption of research into its foreign origins, whether inIndia, Sri Lanka or mainland Southeast Asia.

Genealogical, historical, and political considerations

Arab traders sailing through Southeast Asian waters during the 8th and 9th centuries(CE) reported the existence of a powerful kingdom that held sway over the islands ofthe Malaysian archipelago and occasionally sent off punitive expeditions againstcountries in mainland Southeast Asia. In Arab sources, this paramount kingdomwas known by the name of Za-bag (also transcribed as Za-baj or Zabedj), and itsrulers by their title of maha-ra-ja.

Writing in 1845, and basing himself on some of these Arab reports, J.T.Reinaud thought Zabedj to be centred in the island of Java and to have ruledover various minor kingdoms in Sumatra and the Malay Peninsula (see Reinaud1845, I: lxxiii-lxxv).1 This representation was reversed when G. Coedes, in hisclassic article Le Royaume de Crı-vijaya (1918), had Za-bag equated with Srı-wijaya atPalembang, identifying its rulers with the Sailendras who took special pride incarrying the maha-ra-ja title. Endorsing Coedes’s views, J. Ph. Vogel (1919) andN.J. Krom (1919) independently of each other put forward the idea that Srı--wijaya had extended its hegemony over Java, basing their theories on the Kalasaninscription of 778 and the Kelurak inscription of 782. These texts mention thefoundation in central Java of Buddhist temples on behalf of Sailendra rulers. In

1O.W. Wolters (1979: 1, n.3) suggests that Reinaud explicitly mentioned the kingdom ofSarbaza, i.e. Srı-wijaya, but this is not the case, at least not in the pages specified by Wolters.However, since Reinaud had Zabedj comprising the islands of Sumatra and Java, it would of neces-sity have included Srı-wijaya. Indeed, more recent studies of the Arab reports confirm that someauthors had Sribuza (Srı-wijaya) and Kala-h[-ba-r] (Kedah) listed among the vassal kingdoms of themaha-ra-ja of Za-bag (see, for instance, Tibbetts 1979: 33, 107, 111).

Indonesia and the Malay World Vol. 34, No. 98 March 2006, pp. 3–22

ISSN 1363-9811 print/ISSN 1469-8382 online # 2006 Editors, Indonesia and the Malay World

http://www.tandf.co.uk/journals DOI: 10.1080/13639810600650711

Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

support of his theory, Vogel referred to the Kota Kapur inscription of 686 mention-ing a punitive expedition by Srı-wijayan forces against Java, ‘the land which had notyet submitted to Srı-wijaya’, which appeared consonant with other inscriptionalinformation on the kingdom’s expansion in Sumatra and the Malay Peninsula.Krom had the same expedition against Java linked with the report in Chineseannals about the transfer of the Javanese capital ‘to the East’, somewhere in theperiod 742–775.

However, as a consequence of the discovery, in 1921, of the Na-landa- inscription(Shastri 1924), and following a spurious Dutch debate over the question of whetherthere had been a ‘Sumatran period in Javanese history’ (Krom 1919), or the opposite,namely a ‘Javanese period in Sumatran history’ (Stutterheim 1929), it became clearthat the connection between Srı-wijaya and the Sailendras was not as close as wasassumed. Taking the Sailendras as scions of a northeastern Indian dynasty, R.C.Majumdar (1933) adduced inscriptional evidence to show that they had ruled overJava and over Sumatra, implying that there had been a Sailendra period in both.He also observed that the presence of the Sailendras in Sumatra could not be attestedwith certainty before the 11th century.

With Coedes (1934) conceding to their separation from Srı-wijaya, the positionof the Sailendras in central Java became more puzzling than ever, which resulted ina welter of theories on their possible origins, ranging from various parts of India,to Cambodia, and central Java itself. This mystery has yet to be solved, althoughthe present popularity of the theory of their Javanese origin could suggestotherwise.

It is extremely difficult to follow the erratic course of the debate, not onlybecause of the rapid turnover of different theories but also because of the fact thatalmost all of the leading scholars in the field (Coedes, Stutterheim, Bosch, and DeCasparis) had at one point or another revised their interpretations of the everscarce and ambiguous inscriptional data, inducing them to switch from one theoreti-cal position to another, sometimes even reverting to an earlier point of view (as didBosch, for instance, with respect to the interpretation of the Kalasan inscription, andthe origin of the Sailendra dynasty). It is not surprising, therefore, to find modernauthors, apparently not familiar with all the ‘ins and outs’ of the protracteddebate, inadvertently putting forward ideas that have long since been rejected(e.g. Snellgrove 2000; Hanafiah 2002; Totton 2003). Another complicating factoris that the debate has occasionally been infected by nationalistic sentiments and intel-lectual (self)-censorship. Well-known is the emotional outburst of Poerbatjaraka thatthe theory about the subordinated role of the Javanese ruler Rakai Panangkaran in theconstruction of the Ta-ra- temple at Kalasan reflects a common prejudice about thecompliant nature of the Javanese people.2 Another renowned Old Java hand,Soekmono, was led into the nonsensical statement that the builders of the Hindu-Buddhist temples in Java must have been Javanese because modern Indonesian visitorsto these monuments feel an affinity with them.3 Bambang Sumadio, for his part, whilebeing open to the possibility that the Sailendras were of Sumatran origin, apparently

2‘Mijn sentiment in dit geval niet kunnende bedwingen roep ik uit: wat een belediging!’t Is alsofde Javaan ten allen tijde tot niets anders in staat is geweest dan om ge-“printa”d te worden dooreen vreemde overheerser’ (Poerbatjaraka 1958: 262).

4 I NDONES I A AND THE MA LAY WOR LD

Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

found it more important that their Indonesian ancestry suits the writing of theNational History of Indonesia.4 The fact that some of his compatriots were not yetconvinced of their Indonesian origin, prompted the epigrapher Boechari to accusehis dissident colleagues of ‘enchantment’ with western theories.5 These examplesshow how sensitively Javanese intellectuals can react to what they perceive as anaffront to their national pride and self-esteem.6

In several publications I have argued that the current theory of the Javaneseorigin of the Sailendras is ill-founded, and that there is enough information,however indirect and circumstantial it still may appear, to question the Javaneseor other ‘Indonesian’ origin (whatever is meant by this anachronistic designation).The strongest objection that can be levelled against their Javanese origin is that noone has as yet succeeded in identifying one of the extant names of Sailendra kingswith any of the Javanese rulers listed in the Mantyasih I and Wanua Tengah III inscrip-tions. This finding confirms an early remark by F.D.K. Bosch (1952a: 114) that there-interpretation of the Kalasan inscription by F.H. van Naerssen (1947), whichargued for the involvement of two kings in the construction of the Ta-ra- temple,has much to commend itself as it relieves us from the impossible task of identifyingthe rulers from both series.

The discovery, in 1983, of the list of kings in the Wanua Tengah III inscriptionhas opened a new epoch in the archaeology and art history of central Java, because itcontains the names of all the central Javanese kings ruling in the period between the8th-10th centuries, leaving no breaks and interruptions unaccounted for. In view ofthe ‘complete’ character of the enumeration, the proponents of the single-dynastyhypothesis, who claim that the Sailendras were a native Javanese royal family,should now be held capable of identifying the names of Sailendra maha-ra-jas withrulers listed in the Wanua Tengah III inscription. This is precisely what Boechari(1989, 1990), Kusen (1988, 1994), Jan Wisseman Christie (2001), and Jeffrey Sund-berg (2003) claim to do. However, scrutiny of their work has revealed variousinvalidating flaws in their historical analysis, thereby underscoring – albeit

3Noteworthy is Soekmono’s elaboration on the assumption ‘that the modern Indonesians aredirectly descended from the builders of the candi, and are therefore heirs to these remains ofthe noble works of their ancestors. Such a conclusion is hard to disagree with: the candi stilloccupy a special place in the hearts of the Indonesian people. Their value is difficult to explicate,for it is not a rational, normative quality but an emotional and spiritual one, and can only be trulyfelt by the heirs themselves’ (1993: 51).4Following Boechari’s analysis of the Sojomerto inscription, Sumadio claims that it cannot bedoubted that the Selendra dignitary mentioned therein is an Indonesianisation of the name Sailen-dra. He continues: ‘Persoalannya sekarang yalah apakah Dapunta Selendra itu asli berasal dari Jawaatau Sumatra, kemudian adanya kenyataan bahwa yang penting penulisan Sejarah Nasional Indone-sia, cukuplah untuk membuktikan bahwa ia adalah penduduk asli Indonesia, bukan pelarian atauperebut kekuasan dari luar negri’ (Sumadio 1975: 81, n.28).5‘Teori tentang adanya dua dinasti dilancarkan oleh ahli-ahli Barat [. . .] seperti juga dalam bidang-bidang lain para ilmuwan kita masih terpersona oleh teori-teori bangsa asing’ (Boechari 1989: 1).6Personally, I find this reaction easier to understand than the reluctance of some western archae-ologists and art historians to properly discuss ideas that were introduced into their field of studyby relative outsiders (for particulars, see Jordaan 2000, 2003b).

WHY THE S A I L E NDRAS WERE NOT A JAVANES E DYNASTY 5

Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

indirectly – the rival theory that the Sailendras were a foreign dynasty (see Jordaan2003a; Jordaan and Colless 2004).7

The other reason why I think that the Sailendras were of foreign origin and not aseparate dynasty from another part of the country, is that the establishment of theirrule in Java was accompanied by a number of exogenous changes. In earlier publi-cations (Jordaan 1999a, 1999b), I mentioned the introduction of a new script thatin Dutch colonial times was generally known by the name of Pre-Na-garı- (siddhama--tr.ka), the earliest issuance of the silver Sandalwood-Flower coins, bearing legendsin the same script, the introduction of the maha-ra-ja title and its subsequent adoptionby Javanese rulers, the transfer of the Javanese capital ‘to the East’ (not necessarily toEast Java), and the sudden blossoming of Maha-ya-na Buddhist architectural art. In con-trast, the departure of the Sailendras from Java was followed by such developments asthe fall of Buddhism from royal favour as reflected in the disparaging remarks aboutBuddhist monks and nuns in the Old Javanese Ra-ma-yan.a as well as the halt to Buddhisttemple-building activities, the change from Sanskrit to Old Javanese, the shift fromsilver coinage to an indigenous gold currency. In my estimation, the impact of theSailendras’ departure was so great as to be a major factor in the art-historicalbreak that can be discerned in the temple art of Java (see Jordaan 1999b: 235–39,Jordaan 2003b).8

Similar synchronic changes occurred in other parts of the Indo-Malay archipe-lago, which are analysed in detail in a forthcoming book entitled The Maha-ra-jas ofthe Isles (Jordaan and Colless, in press). For the present purpose, it may suffice torecall W.F. Stutterheim’s (1929) analysis of the curious alternations in the patternof tributary missions to China, with embassies from Srı-wijaya being halted in 742and replaced by missions from Java. However, whereas Stutterheim took this as evi-dence for a Javanese hegemony over Sumatra, we are inclined to relate this to theoverlordship of the Sailendras in the archipelago. One of the indications is thatthe missions from Java were not dispatched from She-p’o (which was the oldChinese designation for Java) but from Ho-ling (thought to be a transcription ofWalaing, a toponym that is connected with the Ratu Boko plateau, which was the

7Presented at the International Conference on Indonesian Art, in New Delhi, 4–6 March 2003,my paper is due to be published in the proceedings of the conference, but copies can be obtainedthrough the KITLV library in Leiden. The final version of the paper has been updated with a post-script in which I take account of Jeffrey Sundberg’s (2003) genealogical identifications in a BKI-article dealing with a Ratu Boko mantra. In the meantime, an enlarged version of the postscripthas been published separately in the Indonesian archaeological journal Berkala Arkeologi (Jordaanand Colless 2004).8Architectural and stylistic changes in temple construction and the plastic arts tend to be morevisible and therefore easier to trace than changes in philosophy and literature, the determinationof which I must leave to experts in these particular fields. Ultimately, they will have to decidewhether my remarks about a possible connection between the traumatic historical event of theeviction of the Sailendras and the comparatively early transfer of literary functions from Sanskritto Old Javanese (see Braginsky 1993: 16) has any factual basis or not. The same holds for mymore speculative linking of this event with the growing preference of the Javanese for theMaha-bha-rata over the Ra-ma-yan. a, and their apparent uneasiness with the performance of lakonfrom the Bha-ratayuddha (see C.C. Berg 1938: 53).


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

site of a famous Buddhist monastery). In contrast, the decline of Sailendra power led toa resumption of missions from She-p’o, while the eviction of the Sailendra from Javaand their settlement in the western part of the archipelago was immediately followedby a resumption of tributary missions from Sumatra, at first hailing from Chan-pei(Jambi), in 853 and 871, and thereafter from San-fo-ch’i. We believe that thistoponym was the Chinese name for the newly reconstituted Sailendra kingdom,and did not refer to Srı-wijaya of old, as is commonly assumed.9 Henceforth, Javaand Sumatra (the latter being ruled by scions of the Sailendra dynasty) were vyingwith each other for recognition by the Chinese court as the pre-eminent kingdomin the region. In the Indo-Malay archipelago itself their formerly mutually beneficialpolitical and economic relationship gave way to hostilities and war.

Socio-cultural considerations

As a detailed discussion of the historical evidence regarding the political alliancebetween Java and Sumatra under the aegis of the Sailendra dynasty goes beyondthe scope of this paper, I will now focus on arguments more specifically related toJavanese culture and society. Hopefully, these considerations will add credence tomy thesis that the Sailendras were a foreign dynasty.

First let me repeat the remark that both before and after the Sailendras the Java-nese must have been familiar with the presence of ‘stranger-kings’, if we can rely onthe fact that the Javanese have special terms for such persons, namely raja sabrang(kings from overseas) and kalana (a wandering adventurer of noble birth fromabroad).10 Examples of stranger-kings in Java include the legendary Aji Saka andrepresentatives of the VOC. In some Javanese mythological texts, such as the SeratKanda [Book of Stories], the Dutch East India Company is classed as a foreigndynasty which managed to gain ascendency in Java by capturing a sick Javanese prin-cess; a story which is strongly reminiscent of the mythical marriage of successiveJavanese sovereigns with ‘king-maker’ Nyai Loro Kidul alias Ratu Kidul, who isstill venerated by numerous present-day Javanese as the goddess of the SouthernOcean. I will return to this shortly.

9In Arab reports we can detect a similar change in nomenclature, in which the names of Ja-baand Za-bag at first stood for the island of Java (as in the earliest reports by Ibn Kurda-dbhih andAbu- Zaid), but following the move of the Sailendras they came to represent Sumatra and poss-ibly also the Malay Peninsula. It must be admitted, however, that the Arab reports are confus-ing, largely because of the indirect and incomplete transmission of the information whichprevented Arab writers in the Middle East from keeping pace with the important politicalchanges occurring in the 9th and 10th centuries (for further details, see Jordaan and Colless.in press).10Long ago, Berg (1929: 12) made the same observation, but found the claims of an Indian ances-try by some East Javanese rulers unacceptable in their present form and without further evidence.Interestingly, Dr David Henley of the Royal Netherlands Institute of Southeast Asian and Carib-bean Studies (KITLV) in Leiden is presently organising an international workshop specificallydevoted to the position of stranger-kings, also in Indonesian history.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

In the same article where I mentioned these (and other Indonesian) examples, Ihave argued extensively that the ‘weak’ version of the so-called ks.atriya theory, whichproceeds from the assumption of a small group of migrants, namely a ruler and hisretinue, instead of mass colonisation (as in the ‘strong’ version of the theory), wasprematurely rejected. I showed that Bosch’ alternative theory about the role ofnative pilgrims or ‘clerics’ in the development of Hindu-Javanese architecture isunsatisfactory in several respects. The most important shortcoming, in myopinion, is that it fails to explain how the Javanese pilgrims, to whom Bosch hadattributed the formidable task of reassembling and unifying the elements fromdiverse Indian architectural traditions, had as individuals managed to achieve thearchitectural and stylistic unity in the Hindu-Buddhist monuments built under theSailendras. Tellingly, Bosch himself deemed this a nearly unsolvable mystery(1952b: 22, 25). My comment was that by focusing on the individual contributionof pilgrims Bosch’s theory incurs a sociological flaw, for:

How could individuals have achieved the alleged unity in Central Javanese archi-tecture, considering their differences in personality, background, and experienceabroad? In my opinion, the only unity which an individual pilgrim would havebeen able to recreate from the materials he had collected in India would havebeen in the form of a single temple, or perhaps a few related temples, probablyof limited size and complexity. The re-creative activity of a number of pilgrimswould have resulted in precisely such a mosaic as Bosch would have expected fromvarious groups of immigrants. The Hindu-Buddhist temples built under theSailendras, on the other hand, are so close to each other in time and design(from an architectural, stylistic, and religious point of view), that they couldonly have been the product of a large-scale collective project.

(Jordaan 1999b: 225)

If this unity in Sailendra monumental art is more likely the product of a large-scalecollective project that was initiated and implemented by the foreign Sailendras, wemust keep in mind that they had access to Indian architects and specialised crafts-men.11 Some support for this idea could be derived from the report byJ. Crawfurd (1820: 221) that his Javanese informants were convinced that theBuddha images in central Javanese temples actually represented ‘foreign priests’(pandita sabrang) who were adherents of Buddhism (agama Buda). Crawfurd commen-ted on this that ‘the bare use of this word [agama Buda], however, which is out of thequestion they could invent, and certainly did not borrow from any modern source,may be considered as satisfactory evidence that they were Buddhists’ (1820: 221). Asa matter of fact, one such foreign priest-architect, hailing from Gaud.ı-dvı-pa (i.e.,Gaud.a in Bengal), is mentioned by name in the Kelurak inscription from 782 (seeBosch 1928: 29–30). This prompts me to repeat the well-known fact that the

11Needless to say, my disputing the theory of the Sailendras’ Javanese origin is not meant todetract from the tremendous and indispensable input which the Javanese had in bringing aboutthe most glorious period in their history. Simply put, without their active participation Borobu-dur and Prambanan could not have been built (nor, for that matter, the town-hall of Batavia orDaendels’s postal road in the days of the Dutch).


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

term zaman Buda is still widely used as a general designation for the pre-Islamicperiod in Java. It is quite conceivable, in my opinion, that the term zaman Budahas remained in use as a reference to the ‘ancient period’ precisely because of theclose association between Buddhism and the evicted Sailendras. Quite a number ofscholars have pointed to the survival of names and toponyms such as Sanjaya,Serayu, Praga, Mataram, from the Hindu-Buddhist period, as illustrative of howthe memory of these long past times has been kept alive.

Concerning the construction of central Javanese temples, and, in particular, theBuddhist structures/monuments, brief reference must be made to the Kalasan inscrip-tion (778) which mentions the fact of the compliance of a Javanese ruler, RakaiPanangkaran, with the request of the guru of a Sailendra king (whose name is notrecorded) to build the Ta-ra- temple at Kalasan. Without going into the philologicalarguments that led van Naerssen (1947) to his reading of the inscription (see alsoBosch 1952a: 113, n.4), but which have been hotly contested by the proponents ofthe single dynasty thesis, I will confine myself to the question of why the buildingproject would have needed the involvement of two rulers in the first place. Van Naers-sen observes that as foreigners the Sailendras were not entitled to dispose of local landand labour, the necessary assets of great works such as the construction of CandiKalasan. To acquire these they were dependent on the indulgence of Rakai Panang-karan. Supporting this interpretation is the information provided by the bilingualcharter from Karangtengah (Kayumvungan), dating to 824, which also alludes tothe co-operation of two parties in the foundation of another Buddhist temple, consist-ing of the Sailendra king Samaratungga and his daughter Pra-modavardhanı-, on the onehand, and Rakai Patapa-n on the other (see de Casparis 1950: 24–50, 105–9). Scholarswho proceed from the assumption that the Sailendras were a native Javanese royalfamily are generally at a loss to explain Samaratungga’s dependence on Rakai Patapa-nfor the donation of the rice-fields for the upkeep of the foundation. For instance,Kusen (1994) speculates that Samaratungga was the son of Rakai Panangkaran fromhis alleged marriage with a Srı-wijayan princess. This hypothetical marriage also pro-vides the basis for his supposition that Samaratungga had never ruled in Java, and thatthis was the reason why his name is not included in the lists of kings in the Mantyasih Iand Wanua Tengah III inscriptions. Instead, Samaratungga would have ruled inSumatra. Kusen finds his theory confirmed in the Kayumvungan inscription, inwhich Samaratungga is interpreted as a foreigner without rights to land. He mightinstead have procured the land through the Javanese wife of Rakai Patapa-n, whomKusen takes for a Sumatran relative. In my earlier mentioned review of Kusen’swork, I have demonstrated that Kusen’s theory is flawed by dubious presuppositions.An illustrative example is his decision to rob Samaratungga of his Javanese nationalityin spite of his alleged descent from a Javanese king, Rakai Panangkaran, and to makehim a foreigner in the land of his forebears without any legal access to the ancestrallands as a consequence of his father’s supposed marriage to a Sumatran princess.More seriously, Kusen fails to adduce evidence for Rakai Panangkaran’s marriage tothis princess, let alone of Samaratungga being the offspring of this union.

Another thing that can be brought to bear on the thesis of the non-Javanese originof the Sailendras is their eviction from Javanese soil. From what I know of the OldJavanese literature, the eviction itself seems difficult to square with what wascommon in Java in ancient times. In the face of an impending military defeat the


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

losing king basically had two options: either to fight to the death or surrender uncon-ditionally.12 In both cases, all his possessions, including his wives and magical objects,fell to the victor.13 The latter was free to decide to kill his vanquished opponent, butas often as not left his opponent on his throne (presumably because of the existence ofmarital and kinship ties between various royal families), provided that the vanquishedking accepted a subordinated position as vassal. Vassals had to bear witness to theirinferior position by paying regular tributary visits to the court of the paramountking. Seen from this perspective, the eviction of the Sailendras from the island ofJava is rather strange and in need of explanation. This provides yet another reasonfor me to think that they were foreigners.

On the other hand, some scholars have hinted at the possibility that the eviction hadsomething to do with religious tensions and conflict: putting the Sailendras, as staunchBuddhists, in opposition to the majority of the population who adhered to Saivism.However, while the religious division itself seems undeniable, the derivative expla-nation is not convincing given that religious differences in ancient Java, in striking con-trast to Europe, were far more common and tolerated, with many rulers extendingliberal support to projects of different denominations. This was the case, for instance,with Rakai Panangkaran, who adhered to Saivism himself but also supported the Bud-dhist cause.With respect to religious differences, Krom and other early Dutch scholarsprobably were biased by what was customary in their native country at the time (i.e. theperiod of Verzuiling), leading them to falsely represent Borobudur and Prambanan asrival monuments. My claim is that the construction of Prambanan began when theSailendras were still in Java and that the building took place with their support,which fits the peaceful co-existence of religions prevailing in ancient Java. If, in spiteof all this, Buddhism came to lose its dominant position to Hinduism and its adherentssuffer from ridicule (as in the Old Javanese Ra-ma-yan.a, mentioned above), this should, inmy opinion, at least partly be explained from the circumstance that the creed wasclosely associated with the foreign Sailendras, whose political influence was decliningduring the first half of the 9th century. Support for this idea can be found in J.G. deCasparis’ (1956: 318, n. 20) interpretation of the Sivagr.ha stone inscription, dated to856, where the revival of Hinduism is represented in terms of a national liberation.

12There is a possible third alternative, namely seeking safety by taking flight, but it seems to methat this was a way for the common people to escape from oppression rather than for royaltyseeking to rebuild their power from scratch elsewhere in Java (not abroad, as did the Sailendras).Moreover, whether such a rare ruler on the run would succeed in the attempt would be depen-dent on the presence of an external party willing to extend its help and protection, as was the casewith Amangkurat II who managed to involve the VOC in his unfortunate war of succession. Thefate of two Madurese princes, who in the days of Sultan Agung had fled to Palembang, was lessfortunate: they were extradited and subsequently killed at Agung’s court. I am indebted toDavid Henley for bringing Amangkurat’s flight to my attention.13Undoubtedly, the capture of the wives and daughters (hanang) was important because of theirsymbolic connection with the fertility of the realm. The magical objects (sacred daggers, rarestones, etc.), extraordinary animals and persons (including albinos, dwarfs, etc.) were alsoseen as a source of supernatural power (kesakten) of the king. Logically or rather mytho-logicallyspeaking, the vassal’s duty to attend to the court of his overlord ( jumurung) indirectly contributedto the latter’s supernatural power (see Berg 1955b: 269, n.74).

1 0 I NDONES I A AND THE MA LAY WOR LD

Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

It is interesting to see what other developments took place in Java after theSailendras’ departure from the island. Remarkable as the decline of Buddhism andthe historical break in temple architecture and art was, no less remarkable wasthe disappearance of the name of the dynasty. To the best of my knowledge, noneof their Javanese successors has ever made a reference to the Sailendras.14 If theSailendras really were a Javanese royal family, it stands to reason that at leastsome of the central Javanese kings from the second half of the 9th century in oneway or another (directly or indirectly, truthfully or fictitiously) would have triedto legitimise their position by tracing their descent from this illustrious family. Ina forthcoming article, Jeffrey Sundberg writes:

It is an enormously difficult question to address how the luster of the name of thisextraordinarily radiant family could have died out so quickly and so thoroughlyafter the issuance of the last known Javanese Sailendra inscription by Samaratung-ga’s daughter in 824. How could the kings of the 830’s and afterwards have failedto claim participation in the name of this dynasty, even when they built theiryounger, smaller, simpler temples in the shadow of the great temples of theSailendra? If the ‘Sanjaya’ were truly Sailendra all along, where did theirfamily name originate and why did the Javanese throne holders abandon iteven while it persisted in Sumatra for at least two centuries?

The ‘persistence’ or rather the continuation of the dynastic name in Sumatra andthe Malay Peninsula is a significant fact which raises a number of questions, eventhough some can only be answered tentatively. For instance, if for the sake of argu-ment we assume that the Sailendras were a Javanese dynasty, why is it that the Java-nese people in the period between February 887 and November 894, which isreferred to in the Wanua Tengah III inscription as ‘leaderless’ (ana-yaka-), did notinvite the Sailendras to return to their land of origin?15 It is well known that forthe Javanese a situation of ‘leaderlessness’ is generally put on a par with social andpolitical unrest and lawlessness, often leading them to look for ‘strong’ personswhom they hope will restore public order.16 The passing over of the Sailendras is

14In a sense, this also holds for the opposite since we know of no Sailendra ruler who has referredto himself by an indigenous Javanese title such as rakai or dyah, as is the case with the Javanesemaha-ra-jas listed in the Wanua Tengah III inscription (cf. van der Meulen n.d.: 21).15However, Kusen (1988: 16) has found that during this so-called leaderless period there were atleast two rulers issuing inscriptions wherein they referred to themselves as maha-ra-ja. Kusen infersfrom this that no ruler was strong enough to hold the position of overlord, and that during thisperiod Mataram seemingly had split into a number of smaller kingdoms.16For a different view, see Anthony Reid (1998). It should be noted, however, that Javanese folk-tradition attests to the tenacity of this idea of power in, for instance, the popular millenarian expec-tations of Jayabaya and the arrival of ratu adil, a righteous king (Sartono 1984). Also relevant is Rick-lefs’ observation, quoted by Reid (1998: 29), that while there were dozens of kings or princeswarring with each other throughout the pre-colonial period, many of the Javanese sources stillfelt there ‘ought’ to have been one king. It seems likely that this political ideal and the hegemonicclaims of the rulers of Singasari, Majapahit, and the Muslim rulers of Mataram, all find their inspi-ration in the glorious days of the Sailendras and their immediate Hindu Mataram successors.

WHY THE S A I L E NDRAS WERE NOT A JAVANES E DYNASTY 1 1

Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

the more remarkable given the fact that they had left Java only recently, that is to saywithin living memory. Conversely, would it not have been ‘logical’ for the Sailendrasto attempt to profit from the power vacuum on their own initiative by presentingthemselves as restorers of dharma (Law), just as their Javanese opponents had donein 856? For those who think this hypothetical scenario far fetched, I must remindthem of even more speculative theories about Sailendra interventions during theeast Javanese period, such as the destruction of the east Javanese capital (kraton) in1006, and the presumed connection between the name of either the wife or eldestdaughter of King Airlangga, Srı- Sanggra-mavijaya Dharmmaprasa-dottungadewı-, withthat of the 11th-century Sailendra king, Sanggra-mavijayottunggavarman.17

The above brings me to the question of how the Javanese, especially those livingabroad, perceive the relationship with their tempat asal, the place of birth or origin.One answer is to see what present-day Javanese migrants in southern Sumatra canteach us about this matter. Regrettably, a quick and admittedly superficial glanceat the literature shows that this is a little researched subject. Relying on a recentstudy on spontaneous migrant settlements in Indonesia (Charras and Pain 1993), itcan be inferred that the majority of Javanese spontaneous migrants in southernSumatra succeed in adapting themselves to their new environment, seeminglywithout much difficulty. Undoubtedly, this is facilitated by the fact that southernSumatra had been an immigration area from the early decades of the 20th centuryonwards. Quoting from this source, it seems that ‘the population rose from 2.2million in 1930, to a combined level close to 20 million today. The most sensationalexample is provided by the province of Lampung, where [official] transmigrants andtheir descendants account for a quarter of the population, spontaneous migrants60%, and native people 15%’. Apparently, the Javanese have a strong tendency toflock together and to create for themselves ‘little Javas’ abroad. Still, the tieswith their tempat asal are never severed, as is for instance borne out by the over-crowded train and bus-loads of Javanese who return to their villages to celebrateLebaran (Idulfitri) or other important events. Probably, as is the case with the Madur-ese, many Javanese settlers abroad attempt to maintain some sort of connection withtheir place of origin, either by their marrying off of (grand)children to cousins ormore distant kin (known in Madurese as mapolong tolang, ‘gathering or keepingtogether of the bones’), or by remittances and investments, and the like. Nothingis known about the Sailendras with respect to either of these two phenomena. Never-theless, the earlier mentioned disappearance of their name in Java does suggest a breakwith Java that is also at odds with what is common among present-day Javanese. On thepossibility of their flocking together with other Javanese refugees, there are, as far as Iknow, no unmistakably Javanese elements (in language and culture) in Sumatra thatcan be traced to the 9th or 10th century, which could be interpreted as a testimony

17Jordaan and Colless (in press) scrutinise some of these theories, and advance several new argu-ments in support of the presumed connection. Nevertheless, further research is needed to confirmthe kinship relation between Sanggra-mawijaya and the Sailendra king Sanggra-mavijayotunggavar-man. This matter has also a bearing on Supomo’s (1972) claim that the use of the designation‘Lord of the Mountains’ in some east Javanese texts should primarily be understood as a symbolicidentification of a supreme king with a national mountain deity rather than as a reference to thefamous dynasty of 8th century central Java, as posited by Berg (Jordaan, forthcoming).


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

of the Javanese origin of the Sailendras. Whatever can be brought to bear on this issueis much more easily explained as a derivation from Indian culture rather than from theJavanese, as is the case with ancient toponyms such as Mataram, and royal customssuch as storage of bricks of gold in a lake (see Balasubrahmanian 1935).

Conclusion and some suggestions for further research

This paper offers several new ideas with respect to the unsolved problem of the originof the Sailendra dynasty, but the main arguments are basically the same as those putforward in other contexts, be it in a different way and for different purposes.While considering the theory of their Javanese origin unlikely, I am well awarethat much remains to be investigated about the Sailendras, and that it would befoolish to think that the Wanua Tengah III inscription has solved all problems ofinterpretation regarding the dynastic relations during their reign in Java. Indeed, asfor the Sailendras, it is not without reason that they were once referred to as an‘evasive race’ (Shastri 1924: 312). The unsolved questions are not just about theircountry of origin, but also on how the Sailendras came to settle in the Indo-Malayarchipelago setting themselves up as widely respected kings for several hundreds ofyears, reaching the pinnacle of their reign in central Java and ending by vanishingfrom the scene in Sumatra. Continued research is needed on their origins: in theIndian subcontinent, Sri Lanka, and mainland Southeast Asia. Preferably this researchshould be supported by inscriptional evidence, including renewed readings of theknown inscriptions (perhaps enhanced with the aid of new detection techniquessuch as laser scans). As it would not be for the first time in the study of ancient Indo-nesian history that a particular idea was prematurely discarded, it could even proveuseful to reconsider some old and neglected hypotheses. Regarding the ks.atriyatheory, it should noted that we no longer have to assume, as Bosch and his contem-poraries used to do, that the spread of Indian influence abroad took the form of a mili-tary conquest and territorial annexation.

As for their settlement in the Indo-Malay archipelago, I will now indicate howthe Sailendras’ position as ‘stranger-kings’ could be further clarified by comparativehistorical research along the lines explored by Marshall Sahlins (for the Pacific region)and David Henley (on northeastern Sulawesi). Sahlins’ (1981) article is especiallyuseful for his discussion of the complex conceptual linkages between stranger-kings and the native peoples of Hawaii, Fiji and other Pacific islands, which are com-monly interpreted and represented in gendered terms. As Sahlins remarks:

The [stranger-]king is an outsider, often an immigrant warrior prince whosefather is a god or a king of his native land. But, exiled by his own love ofpower or banished for a murder, the hero is unable to succeed there. Instead,he takes power in another place, and through a woman: princess of the nativepeople whom he gains by a miraculous exploit involving feats of strength,ruse, rape, athletic prowess and/or murder of his predecessor. The heroicson-in-law from a foreign land demonstrates his divine gifts, wins the daughter,and inherits half or more of the kingdom.

(Sahlins 1981: 115, emphasis in the original)


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

Those familiar with the Javanese Panji stories will recognise the similaritiesbetween these Pacific exploits and the equally miraculous prize contests in ancientJava that were known by the term sayembara. Unlike the Indonesian sayembara oftoday, which usually pivot round material objects and money, the ultimate prizein ancient Java invariably was a royal princess, and marrying her was a sure way togain access to the land. Panji is the name of the Javanese culture hero who defeatshis rivals (including several kalana kings) before marrying his true love.

In other Old Javanese stories, like the earlier mentioned Serat Kanda, it is theVOC’s capture of a sick Javanese princess that served to explain the ascendancy ofthe Dutch in Java. The sick princess, who either suffers from a skin disease or hasa flaming womb, returns me to my early research on Nyai Lara Kidul alias RatuKidul (Jordaan 1984, 1987, 1997), to whom – to paraphrase my late friend HanResink – I have remained deeply attached. The sick princess, in my opinion,represents none other than the primeval fertility goddess of the Javanese, who,being a vacillating goddess, under Hindu influences came to be identified bothwith the awesome goddesses like Durga- and Ka-lı- (her demonic or malevolent face)and Srı- Devı- or similar benevolent goddesses like Uma-, Pa-rvatı-, and Laks.mı- (her ben-evolent face).18 Without going into the complex transformations in her character andappearance, the capture or possession of the sick princess was a well-known mytho-logical motif to explain the rise to power of her lover and/or abductor, who as a ruleis remembered as the founder of a new dynasty.19 What matters here is that real his-torical events are interpreted from this indigenous mythological perspective, andthat historical personages are thus associated or identified with particular (semi-)divine beings. This, as Sahlins (1981: 107–9) explains, is what happened with themurder of Captain Cook by the Hawaiians in 1779. The mythological interpretationof this dramatic event found verbal expression by the islanders when they brought apiece of Cook’s hindquarters to his ship, the Resolution, asking the crew when Lono,the god of agriculture with whom Cook apparently had been identified, would comeback to them. As for pre-colonial Java, Berg has shown how the Javanese court poetstried to make sense of the military defeat of Sultan Agung in 1629 at the hands of theVOC by ‘inventing’ the story of the (temporary) capture of the sick princess. Inmuch the same way, I assume, the Javanese would have explained the ascendancyof the Sailendras from a mythical marriage or a real marriage cloaked in mythological

18Gradually, however, the two aspects of the autochthonous fertility goddess were loosened andseparated, with the darker side of Nyai Lara Kidul being associated with the fierce and heavilypoliticised Durga- (thus underscoring her position of ‘king-maker’ and protector of the realm),while her benevolent side merged with the rather dull and heavily domesticated ‘Dewi Sri’(who became the goddess of rice for the Javanese). Only in special circumstances do these disso-ciated goddesses reveal something of the suppressed side in their characters. Given their primevalunity, Nyai Lara Kidul and Dewi Sri are often attributed with the same chthonic attributes infolklore, such as a skin disease and/or a snake-like appearance. In this regard, C.C. Berg(1955a: 375) was not wide of the mark in referring to Nyai Lara Kidul as ‘the maritimeversion of Dewi Sri’.19The better-known examples are Ken Angrok, Raden Susuruh, and Senapati. For more particu-lars, see Pigeaud (1927), Berg (1938), Jordaan (1984, 1987, 1997), and Wessing (1977a, 1977b),among others.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

garb with a Javanese royal princess. Perhaps it was this marriage that resulted in adivision of central Java, the faint echoes of which might have reached the Chineserecords about the move of the Javanese capital to the East during the t’ien-paoperiod (742–755) and also the subsequent substitution of diplomatic missions ofShe’po by those from Ho-ling. More significantly, in one of the three Ratu Bokoinscriptions which were issued after (and presumably in commemoration of) thedefeat of the Sailendras explicit reference is being made to Laks.mı- ‘who bearsMajesty necessarily hidden in the juncture of her legs’. This strange expression, asDe Casparis (1956: 266) observed, brings to mind the flaming-wombed princessKen D.ed.es, whose possession enables the usurper Ken Angrok to become king ofeast Java and gain control over the fertility of the realm.20

Mindful of the importance of the dictum cherchez la femme, I have often looked forevidence of a politico-religious association between the historical Ta-ra- of theNa-landa- inscription, the Buddhist goddess Ta-ra-, to whom, it may be recalled, theSailendra temple at Kalasan was dedicated, and Nyai Lara Kidul.21 As was tobe expected given the scarcity of historical information, my efforts met with littlesuccess, although enough art historical data came to light to identify Vyasa-Ta-ra- asthe presiding goddess of the Kalasan temple (Jordaan 1997).

The extant historical information is and will remain scarce and fragmentary,forcing us to open up other sources. In this context, it needs to be rememberedthat there is at least one ancient literary text, namely the Ra-ma-yan.a Kakawin, thatcould be tapped for additional, if indirect and veiled information on the Sailendrasand, perhaps, Nyai Lara Kidul.22 Research by W. Aichele (1969) has confirmed Poer-batjaraka’s (1927) early surmise that some parts of the Old Javanese Ra-ma-yan.a could

20De Casparis adds the following remark: ‘With a view to these demonic and sensual aspects ofCaivism, it may not be superfluous to stress that it is now beyond doubt that they were knownduring the greater part of Hindu-Javanese history’ (1956: 266). Further research is needed toestablish whether the motif of the flaming womb (in contradistinction from skin disease,which seems of genuinely Austronesian origin) was derived from Tantrism, which must havereached Java much earlier than was generally assumed before World War II. As De Casparisnotes, ‘All the material at our disposal seems to indicate a co-existence of many differentforms of religion and worship during Old Javanese history’ (1956: 266, n.99).21Stutterheim (1929) has tried to relate the construction of the Ta-ra- temple at Kalasan with themarriage of the Sailendra king Samaragra-vı-ra to Princess Ta-ra- who is mentioned in the Na-landa-

inscription, but his proposal was rightly rejected as chronologically untenable (Bosch 1929).22My theory is that the Sailendras were instrumental in the association of this autochtonous fertilitygoddess with deities from theHindu-Buddhist pantheon and also in the transformation of this Javanesefolk deity into a Nagı- royal consort (see also Gaudes 1993, for a comparison with Cambodia). From acultural and historical point, it is unnecessary to assume, as did RobertWessing (1977b: 326), that itwas through their military raids against mainland Southeast Asia during the 8th century that the Java-nese came ‘into contact with Cambodian court culture, with its tale of the ruler’s nightly cohabita-tion with the source of his realm’s prosperity’. In whatever way this process of Indianisation/Sanskritisation took place in central Java and who was involved in it, I think that Wessing willagree that the ideas underlying these Indianised customs were known throughout monsoon Asiafrom time immemorial – as was for instance the case with the various forms of construction sacrificewhich were the subject of our joint publication (e.g. Wessing and Jordaan 1997).


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

date from the late central Javanese period since they contain all kinds of metaphorichints to Rakai Pikatan (reign period 847–858) and his alleged deception by Buddhistdignitaries. Berg (1955: 253) was even inclined to assume that the text was written incommemoration of the destruction of the Sailendra dynasty, and that the Watugu-nung story in the Badad tanah Jawi was a folk myth that had ‘accompanied’ the officialRa-ma-yan.a. Those familiar with the work of Berg and his unsurpassed talents to seethrough all kinds of clever Old Javanese magico-literary plots and identifications,will be unsurprised to learn of his equating Ra-van.a with the Sailendras, andHanuman (The White Monkey) with Vis.n.u. If indeed there is a mythico-historicalbasis for linking the Sailendras with Ra-van.a (who after all is a raja sabrang), thereis a likelihood of his meeting Sı-ta- as a sick princess during her captivity in Lan.ka-.A renewed and careful study of these ancient texts is therefore needed to establishthe plausibility of these and other identifications. A new critical English edition ofthe Old Javanese Ra-ma-yan.a would be meritorious in itself, quite apart from its use-fulness as an alternative source of historical information.

Henley’s analysis seems more useful for comparative research on the local socio-economic and political conditions that might have facilitated the acceptance of theSailendras’ authority by the Sumatran peoples at the end of the 7th century.Briefly summarised, Henley’s thesis is that the relatively easy introduction ofDutch rule in northern Sulawesi cannot be explained from an alleged Dutch super-iority in military power or their cunning application of the strategy of divide andrule, but rather from the inability of the local peoples and their rulers to put anend to their ongoing internecine conflicts and warfare. As Henley demonstrates,these internal conflicts and the peoples’ deep-seated mutual suspicion and jealousy,increasingly made them turn to the Dutch, who, being outsiders, were asked to actand/or intervene as arbiters and impartial judges. Henley (2002: 55) states that thistendency was not without consequences: ‘The rise of the colonial state in northernSulawesi, then, can be regarded in part as a deliberately accepted solution to pro-blems of mutual cooperation which were perceived by indigenous actors as difficultto solve without the aid of a powerful and impartial external party’.23 In my opinion,a similar scenario could be envisaged with respect to the ready acceptance of Sailen-dra overlordship in Sumatra and the Malay Peninsula at the end of the 7th century. Inthe forthcoming book, referred to above, it is remarked that

If the family was as illustrious as the Sailendras themselves claimed to be, prob-ably descending from a royal family that hailed from India, they had several

23ThoughHenley himself focusses on the socio-political dimension of the stranger-king phenomenon,he indicates that the mythico-religio aspects are also present. This is evident, for instance, from thefact that VOC judgments were regarded as Godsspraken (‘divine verdicts’ or oracles). Also relevant isHenley’s observation that royal status in northern Sulawesi was often framed in an ‘idiom of strange-ness’, as in Gorontalo where the kings were chosen from groups who claimed descent ‘from foreignimmigrants (including characters from the Bugis epic poem I La Galigo), and partly from semi-humancreatures which emerged from eggs, clumps of rattan, or shafts of bamboo long after the originalpeople of the country’ (Henley 2002: 64–65). It is easy to see that such information offers a possi-bility of linking Henley’s analysis with that of Sahlins’, discussed above, which focuses on the culturalassumptions relating to the stranger-king phenomenon.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

things to offer: not territory, but prestige, overseas contacts in India and SriLanka, and knowledge. This knowledge would have covered Indian religion,military science, administration, and statecraft.24 The latter, in particular,would have been a valuable asset considering Srı-wijaya’s problems of socialcontrol and internal political stability. These problems are indicated, forinstance, in the Telaga Batu or Sabokingking inscription, which sets forth along list of potential enemies, ranging from slaves, washermen, merchants,and naval captains to ‘sons of kings’.

Following the interpretations of De Casparis (1956: 15–47), Wolters hasinferred from this seemingly ‘paranoid’ text that ‘evidently the ruler was at thattime concerned with the problem of keeping under control a disturbed realm, poss-ibly including recently conquered territories’ (Wolters 1961: 17; cf. WissemanChristie 1982: 296, 312; 1984: 55). It is hardly conceivable that the Srı-wijayakingdom could have bridged the long period of seven centuries of its presumedexistence on the subtle and limited check-and-balance political mechanisms ofindigenous Sumatran societies discussed in Kenneth Hall’s (1976) study of earlySrı-wijayan inscriptions (cf. Kulke 1991: 15). In our opinion, it was the alliancewith the Sailendras that saved Srı-wijaya’s name from oblivion (Jordaan andColless, in press).

In line with Henley’s analysis, it could be argued that the divisive and unstableconditions apparently prevailing in southern Sumatra during the 7th and 8th centu-ries would have contributed to the acceptance of Sailendra rule by the local commu-nities. As stranger-kings the Sailendra were outside and above these fractious andwarring communities, enabling them to act as mediators and impartial arbitrators.This need for impartiality would have been met by the religious tolerance and het-erodox inclinations for which the Sailendras were known, and the ideological empha-sis they put on the abstract notion of the Word as reflected in their central Javaneseinscriptions. The latter resonates with the surat cap (sealed letters) of Minangkabauroyalty (Drakard 1993), who as we speculate could well have been their distantdescendants.25

Regrettably, we cannot, at the present state of knowledge, say whether similarsocio-political conditions prevailed in central Java that would have facilitated the

24In response to Wisseman Christie’s (1990: 41) critical remarks against such views, we wantto stress that we do not imply that some ‘essential’ ingredient was lacking in the south Suma-tran society in question. Rather than representing a generally negative view, we would arguethat the Indian contributions should be seen as ‘positive’, as enriching influences, quite similarto the cultural impact of the Pax Romana on the development of different West Europeansocieties.25Noteworthy are the remarkable (but still to be investigated) similarities between theSailendras and the former Minangkabau royal family. For instance, they both stood apartfrom the surrounding society by their ‘contrapuntal’ descent and marriage rules, as wellas by their heterodox and syncretic tendencies. Interestingly, the Minangkabau still maintainconnubial relations with Negri Sembilan, on the other side of the Straits of Malacca, whichbrings to mind the (marital) alliance between Kedah and Malayu during the reign of theSailendras.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

establishment of Sailendra rule in this island.26 The only extant document prior tothe arrival of the Sailendras is the Canggal inscription of 732, issued by KingSanjaya in celebration of the foundation of a lin.ga sanctuary, but it is precisely thisinscription, in conjunction with the information derived from an Old Sundanesetext, the Carita Parahyangan, that has been used as evidence of the regional supre-macy of the Javanese, and to defend the identity of the ‘Sanjayas’ and the Sailendras.While we can discard the Carita Parahyangan as a late and unreliable source of his-torical information, we would be wise to accept the integrity of the Wukir inscrip-tion concerning Sanjaya’s stature as a great Javanese king. This would suggest that theestablishment of Sailendra rule in central Java was different from that of Sumatra andthe Malay Peninsula, and that it may well have taken place with Sanjaya’s consent or,much more likely, with that of his son, Rakai Panangkaran (reign 746–784), whowas closely involved in the construction of the Ta-ra- temple at Kalasan and presum-ably the first Javanese king to adopt the maha-ra-ja title. The most plausible way for thiscourse of events to have happened was a marital alliance of the two royal families. Inthe absence of pertinent information, we are left to speculate how this presumedmarital alliance was arranged, allowing for the possibility that the marriage was ofa rather unusual type, for instance with the two kings arranging a sister-exchangemarriage, or practising some form of polyandry (see e.g. Moens 1937: 438–43,1939: 77–79). Comparative research on kinship and marital practices amongancient Indian dynasties could perhaps shed new light on this matter (see e.g.Singh 1978; Chakraborti 1984; Trautmann 1974, 1981). If future research doesconfirm the existence of a marital alliance between the Sailendras and the Javaneseroyal family of Rakai Panangkaran, this clearly will help to explain why the later evic-tion of the Sailendras had such a deep and long traumatic impact on the society andculture of Old Mataram, and Java as a whole.

Acknowledgements

Considering that this is my final statement about the Sailendras in central Javanese history, andhaving them tentatively linked with the Javanese goddess Nyai Lara Kidul whose mysterious char-acter has attracted me for more than 20 years, I am inevitably reminded of the intellectual andmoral support received from various persons over the years. I feel especially indebted to Bertvan den Hoek and Lokesh Chandra for introducing me to the study of ancient Indian historyand culture, and to Han Resink, Hans Teeuw, and Willem van der Molen for welcoming meto ‘their’ field of Javanese studies. Being of Dutch-Madurese extraction and more familiar withrural Madurese people through anthropological fieldwork, I am grateful to these scholars for

26According to Henley (2004: 47) such was the case during the days of the VOC:The proximate reason for the vulnerability of Javanese states to hesitant or reluctant VOCintervention in the seventeenth and eighteenth centuries [. . .] was surely that too manymembers of the Javanese mobility always hated each other more than they hated any foreigner,so that the Company was “inexorably sucked into Javanese affairs” (Nagtegaal 1996: 16) byopportunistic requests for support in civil conflicts until finally a stable equilibrium wasreached in which the Dutch found themselves playing what Ricklefs (1974: 420–1) explicitlycalls “a mediatory role” as “ultimate arbiters of insoluble disputes” between two permanentlyseparate Javanese kingdoms of roughly equal strength.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

helping me to find my way in the court cultures and literature of ancient Java. Fortunately, theyshowed great forbearance towards my non/un-Javanese outspokenness and occasional fits ofemotions (usually over the treatment by other Old Java specialists). Finally, I want to thankBrian Colless, David Henley, and Jeffrey Sundberg for their valuable comments on the draft ofthis paper.

References

Aichele, W. 1969. Vergessene Metaphern als Kriterien der Datierung des altjavanischenRa-ma-yan.a. Oriens Extremus 16: 127–67.

Balasubrahmanian, S.R. 1935. The lake of the bricks of gold. Tijdschrift voor Indische Taal-,Land- en Volkenkunde (TBG) 75: 618–19.

Berg, C.C. 1929. Hoofdlijnen der Javaansche litteratuurgeschiedenis. Groningen/Den Haag,Weltevreden: Wolters.

Berg, C.C. 1938. Javaansche geschiedschrijving. In F.W. Stapel (ed.), Geschiedenis vanNederlandsch Indie. Vol. II. Amsterdam: Joost van den Vondel, pp. 7–148.

Berg, C.C. 1955a. De zin der tweede Babad-Tanah-Jawi. Indonesie 8(1): 361–400.Berg, C.C. 1955b. Geschiedenis en geschiedschrijving vanMataram. Indonesie 8(3): 231–69.Boechari, M. 1989. Satu atau dua dinasti di kerajaan Mataram kuna? Jakarta: FSUI.

Mimeograph.Boechari, M. 1990. Tafsiran atas prasasti Wanua Tengah III. Jakarta: FSUI. Mimeograph.Bosch, F.D.K. 1928. De inscriptie van Keloerak. Tijdschrift voor Indische Taal-, Land- en

Volkenkunde 68: 1–64.Bosch, F.D.K. 1929. Dr W.F. Stutterheim, A Javanese period in Sumatran history, Sur-

akarta 1929. Tijdschrift voor Indische Taal-, Land- en Volkenkunde 69: 135–56.Bosch, F.D.K. 1952a. Crıvijaya, de Cailendra- en Sanjayavamca. Bijdragen tot de Taal-,

Land- en Volkenkunde 108: 113–23.Bosch, F.D.K. 1952b. ‘Local genius’ en Oud-Javaanse kunst. Mededelingen der Koninklijke

Nederlandse Akademie van Wetenschappen, Afdeling Letterkunde, N.S.15–1: 1–25.Braginsky, V.I. 1993. The system of classical Malay literature. Leiden: KITLV Press.

Working Papers 11.Charras, M. and Pain, M. (eds). 1993. Spontaneous settlements in Indonesia; Agricultural

pioneers in Southern Sumatra. Bondy/Jakarta: Orstom-CNRS/DepartemenTransmigrasi.

Casparis, J.G. de. 1950. Inscripties uit de Cailendra-tijd. Bandung: Nix.Casparis, J.G. de. 1956. Prasasti Indonesia II: Selected inscriptions from the 7th to the 9th cen-

turies AD. Bandung: Masa Baru.Chakraborti, Haripada. 1984. Cross-cousin marriage in ancient India. In Religion and

society in ancient India; Sudhakar Chattopadhyaya commemoration volume. Calcutta:Roy & Chowdhury, pp. 398–405.

Coedes, G. 1918. Le royaume de Crıvijaya. Bulletin de l’Ecole Francaise d’Extreme-Orient8:1–36.

Coedes, G. 1934. On the origins of the Sailendras of Indonesia. Journal of the Greater IndiaSociety 1: 61–70

Crawfurd, J. 1820. History of the Indian archipelago containing an account of the manners, arts,languages, religions, institutions, and commerce of its inhabitants. Edinburgh: ArchibaldConstable.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

Drakard, J. 1993. A kingdom of words: Minangkabau sovereignty in Sumatran history.PhD thesis. Australian National University.

Gaudes, R. 1993. Kaundinya, Preah Thaong, and the ‘Nagı Soma’: Some aspects of aCambodian Legend. Asian Folklore Studies 52: 333–58.

Hall, K.R. 1976. State and statecraft in early Srivijaya. In K.R. Hall, and J. K Whitmore,(eds), Explorations in early Southeast Asian history: The origins of Southeast Asian state-craft. Ann Arbor: Center for South and Southeast Asian Studies, University ofMichigan, pp. 61–105.

Hanafiah, Djohan. 2002. Siapakah dinasti Sailendra? In 25 tahun kerjasama Pusat PenelitianArkeologi dan Ecole francaise d’Extreme-Orient. Jakarta: EFEO, pp. 83–97.

Henley, D. 2002. Jealousy and justice. The indigenous roots of colonial rule in northern Sulawesi.Amsterdam: VU Uitgeverij.

Henley, D. 2004. Conflict, justice, and the stranger-king; indigenous roots of colonialrule in Indonesia and elsewhere. Modern Asian Studies 38: 85–144

Jordaan, R.E. 1984. The mystery of Nyai Lara Kidul, Goddess of the Southern Ocean.Archipel 28: 99–116.

Jordaan, R.E. 1987. Skin disease, female ancestry and crops. In E. Locher-Scholten andA. Niehof (eds), Indonesian women in focus. Dordrecht: Foris, pp. 120–34.

Jordaan, R.E. 1997. Ta-ra- and Nyai Lara Kidul: Images of the divine feminine in Java.Asian Folklore Studies 56(2): 255–312.

Jordaan, R.E. 1999a. The Sailendras in central Javanese history; A survey of research from 1950to 1999. Yogyakarta: Universitas Sanata Dharma.

Jordaan, R.E. 1999b. The Sailendras, the status of the ks.atriya-theory, and the develop-ment of Hindu-Buddhist art in Java. Bijdragen tot de Taal-,Land- en Volkenkunde155(2): 210–43.

Jordaan, R.E. 2000. Consensus and variance in Indonesian archaeology; A reply to Prof.John Miksic. Bijdragen tot de Taal-, Land- en Volkenkunde 156(1): 169–75.

Jordaan, R.E. 2003a. The Wanua Tengah III inscription and the problem of the origin ofthe Sailendra dynasty. Paper presented at the International Conference of Indone-sian Art, New Delhi, IGNCA, 3–6 March 2003.

Jordaan, R.E. 2003b. Continuity or discontinuity in Hindu-Buddhist temple art inancient Java? Dialogue 5(1): 93–100.

Jordaan, R.E. Forthcoming. Who was Srı- Sanggra-mawijaya Dharmmaprasa-dotunggadewı-?Jordaan, R.E. and Colless, B.E. 2004. The Ratu Boko mantra and the Sailendras. Berkala

Arkeologi 24(1): 56–65.Jordaan, R.E. and Colless, B.E. In press. The Maha-ra-jas of the isles; The Sailendras and the

problem of Srı-wijaya. Leiden: Semaian.Krom, N.J. 1919. De Sumatraansche periode der Javaansche geschiedenis. Leiden: Brill.

Inaugural address.Kulke, H. 1991. Epigraphical references to the ‘city’ and the ‘state’ in early Indonesia.

Indonesia 52: 3–22.Kusen. 1988. Prasasti Wanua Tengah III, 830 Saka: Studi tentang latar belakang peruba-

han status sawah di Wanua Tengah sejak Rake Panangkaran sampai Rake WatukuraDyah Balitung. Makalah dalam Kegiatan Ilmiah Arkeologi, IAAI Komisariat Yogya-karta–Jawa Tengah, Yogyakarta, 2 September 1988.

Kusen. 1994. Raja-raja Mataram Kuna dari Sanjaya sampai Balitung sebuahrekonstruksi berdasarkan prasasti Wanua Tengah III. Berkala Arkeologi 14: 82–95.Special issue.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

Majumdar, R.C. 1933. Les rois Sailendra de Suvarnadvıpa. Bulletin de l’Ecole Francaised’Extreme-Orient 33: 121–41.

Meulen, W.J. van der. n.d. Pedoman literatur; Sedjarah untuk memperlengkapi bahankuliah: I. Disekitar persoalan Sailendra-vamsa. [n.p.]: IKIP Sanata Dharma–Djurusan Sedjarah. Mimeograph.

Moens, J.L. 1937. Crıvijaya, Yava en Kata-ha. Tijdschrift Tijdschrift voor Indische Taal-,Land- en Volkenkunde (TBG)77: 317–487.

Moens, J.L.1939. Srı-vijaya, Ya-va and Kata-ha. Abridged English translation byR.J. Touche. Journal of the Malayan Branch Royal Asiatic Society 17(2): 1–108.

Naerssen, F.H. van. 1947. The Cailendra interregnum. In India Antiqua. Leiden: Brill,pp. 249–53.

Nagtegaal, L. 1996. Riding the Dutch tiger; The Dutch East Indies Company and the northeastcoast of Java, 1680–1743. Leiden: KILTV Press. VKI 171.

Pigeaud, Th.G.Th. 1927. Alexander, Sakender en Senapati. Djawa 7: 321–62.Poerbatjaraka. 1927. De dateering van het Oud-Jav. Ra-ma-yan.a. In Gedenkschrift uitgegeven

ter gelegenheid van het 75-jarig bestaan op 4 juni 1926 [KITLV].’s-Gravenhage:Nijhoff, pp. 265–72.

Poerbatjaraka. 1958. Crıvijaya, de Cailendra- en de Sanjayavamca. Bijdragen tot de Taal-,Land- en Volkenkunde 114: 254–64.

Reid, A. 1998. Political ‘tradition’ in Indonesia: the one and the many. Asian StudiesReview 22: 23–38.

Reinaud, J.T. 1845. Relations des voyages fait par les Arabes et les Persans dans l’Inde et a laChine. Paris: L’Imprimerie Royal. 2 vols.

Ricklefs, M.C. 1974. Jogjakarta under Sultan Mangkubumi 1749–1792; A history of thedivision of Java. London: Oxford University Press.

Sahlins, M. 1981. The stranger-king or Dumezil among the Fijians. Journal of PacificHistory 16(3): 107–32.

Sartono Kartodirdjo. 1984. Ratu Adil. Jakarta: Sinar Harapan.Snellgrove, D. 2000. Asian commitment; Travels and studies in the Indian sub-continent and

South-East Asia. Bangkok: Orchid Press.Shastri, Hirananda. 1924. The Nalanda copper-plate of Devapaladeva. Epigraphia Indica

17: 310–27.Singh, Sarva Daman. 1978. Polyandry in ancient India. Delhi: Motilal Banarsidass.Soekmono, R. 1993. The candi as a cultural pusaka. In H. Soebadio (ed.), Art of Indonesia.

London: Tauris Parke, pp. 51–57.Stutterheim, W.F. 1929. A Javanese period in Sumatran history. Surakarta: De Bliksem.Sumadio, Bambang (ed.). 1975. Jaman Kuno, Sejarah Nasional Indonesia. Vol. 2. Jakarta:

Departemen Pendidikan dan Kebudayaan.Supomo, S. 1972. ‘Lord of the mountains’ in the fourteenth century Kakawin. Bijdragen

tot de Taal-, Land- en Volkenkunde 128(2/3): 281–297Sundberg, J. 2003. A Buddhist mantra recovered from the Ratu Baka plateau: A prelimi-

nary study of its implications for Sailendra-era Java. Bijdragen tot de Taal-, Land- enVolkenkunde 159(1): 163–88.

Sundberg, J. Forthcoming. The state of the state of Mataram: A review of recent effortsto clarify its history.

Tibbetts, G.R. 1979. A study of the Arabic texts containing material on South-East Asia.Leiden: Brill.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

Totton, M-L. 2003. Narrating animals on the screen of the world. Art Bulletin 85(1):6–25.

Trautmann, Th. R. 1974. Kinship and history in South Asia. Ann Arbor: University ofMichigan.

Trautmann, Th. R. 1981. Dravidian kinship. Cambridge: Cambridge University Press.Vogel, J.Ph. 1919. Het koninkrijk Crıvijaya. Bijdragen tot de Taal-, Land- en Volkenkunde

75: 626–37.Wessing, R. 1977a. Nyai Roro Kidul in Puger: Local applications of a myth. Archipel 53:

97–120.Wessing, R. 1977b. A princess from Sunda: some aspects of Nyai Roro Kidul. Asian Folk-

lore Studies 56(2): 317–53.Wessing, R. and Jordaan, R. E. 1997. Death at the building site: construction sacrifice in

Southeast Asia. History of Religions 37(2): 101–21.Wisseman Christie, J. 1982. Patterns of trade in western Indonesia: Ninth through thir-

teenth centuries AD. PhD thesis. School of Oriental and African Studies, Univer-sity of London.

Wisseman Christie, J. 1984. Trade and early state formation in maritime Southeast Asia:Kedah and Srivijaya. Jebat 13 (1984/85): 43–56.

Wisseman Christie, J. 2001. Revisiting early Mataram. In M.J. Klokke and K.R. vanKooij (eds), Fruits of inspiration; Studies in honour of Prof. J.G. de Casparis. Groningen:Forsten, pp. 25–57.

Wolters, O.W. 1961. Srı-vijayan expansion in the seventh century. Artibus Asiae 24:417–24.

Wolters, O.W. 1979. Studying Srı-vijaya. Journal of the Malaysian Branch Royal AsiaticSociety 52(2): 1–31.


Dow

nloa

ded

by [

Inst

itutio

nal S

ubsc

ript

ion

Acc

ess]

at 1

7:13

01

Sept

embe

r 20

11

Documents

Shailendra origins