Urban images (3) – Cornerstones with protection on Delos

Miko Flohr, 21/02/2020

This is the corner in the outer wall of a second-century BCE house in Delos. You can clearly see that a different type of stone has been used to reinforce the corner, and it is hardly possible to miss how they have been decorated. The phallus played an important apotropaic role in Greco-Roman culture and could be found depicted on many places where it was meant to keep out bad fortune. The mace – if that is what is depicted on the other slab with decoration – could perform a similar role (and, for that reason, could become a symbol for Hercules).

Urban images (2) – Concrete with pebbles in Hadrianic Itálica

Miko Flohr, 14/02/2020

Though the Romans are renowned for their use of concrete, which allowed for quick construction and liberated architects from the constraints imposed by stone-blocks that had dominated Greek and Hellenistic architecture, it should not be overlooked that heavy concrete was and remained very much an Italian phenomenon: it is found throughout the Roman empire in public architecture, but in private architecture, not so much.

One exception is the city of – what’s in a name – Itálica, near Sevilla, in Spain, which is the place from where the families of the emperors Trajan and Hadrian came. Under Hadrian, Itálica was massively extended, and in the new quarter, private construction is heavily dominated by concrete. Yet, if you look carefully, you’ll see that this is not the usual brick-faced concrete that you can see at Ostia, but it is made with pebbles instead, thus adapting a familiar building technique to a new environment.

Global Romans (6) – Fake Roman Coins from the Mekong Delta

Miko Flohr, 11/02/2020

If Ptolemy’s Golden Peninsula could be identified with the Thai-Malay peninsula, the city of Kattigara is bound to remain a mystery: it simply cannot be securely identified on the basis of the evidence we have at our disposal. Of course, this has not kept people from making more (or less) educated guesses. Based on the coordinates given by Ptolemy, a lot is possible.

Some have argued Kattigara must be equated with Saigon in Vietnam. This is not theoretically impossible, but Saigon only emerged as a settlement in the medieval period, and the oldest remains in the area belong to a temple dating to the fourth century CE. Others have linked Kattigara with the archaeological site of Óc Eo, also in Vietnam, but on the other side of the Mekong delta – and thus much closer to Thailand. At least, in this place, objects and structures dating to the first centuries of our era have been dug up, but there is no compelling reason to link these to the place mentioned by Ptolemy. We might as well give up. It is plausible, and perhaps likely, that Ptolemy’s Kattigara was somewhere in the Mekong region, but unless we find a place with the name Kattigara written all over it, this is as far as we will get.

Still, the Mekong delta, and particularly the excavations at Óc Eo, are highly relevant to our understanding of the ties between the Far East on the one hand and the Far West on the other. While the history of the region resembles that of Java in this period in that there are no textual sources from the area, it is clear that, by the time of Ptolemy, some kind of larger-scale polity existed in this place – it is commonly known under its Chinese name ‘Funan’, and seems to have emerged in the first century of our era, and continued to exist until the fifth century, and beyond.

While our understanding of ‘Funan’ remains limited and constrained by the Chinese perspective in the available sources, it is clear that the kingdom – which is what it supposedly was – was in frequent contact with China, and had continuous ties with India. It is also probable that Romans visited the area: both at Óc Eo, and at another high-profile site from the area, Angkor Borei in Cambodia, Roman coins have been found, and at Óc Eo, several other artefacts from the Far West, comparable to those discussed last week, have been discovered.

Pendants imitating Roman Coins from the Funan region

Yet the most surprising ‘Roman’ objects from the kingdom of Funan were not made in the Mediterranean, but in the kingdom itself. In several places in the region, including, but not limited to Óc Eo, locally produced pendants imitating Roman imperial coins have been found. That is to say, local metalworkers took the design of Roman coins, replicated it—not necessarily very precisely—and produced new objects on that basis, which they then could attach to a necklace. Apparently, for people in the region, Roman coins, with their detailed imagery, had a certain intrinsic aesthetic value.

Many people in Funan may not have known anything about the Roman Empire, but some of them happily walked around with Roman coins on their necklaces. They were not the only ones: similar ‘imitations’ can be found in India, and here, too, imitated Roman coins took the form of pendants. It is possible that the people of Funan simply copied something that they had seen in India. However, recent work has shown that the pendants of Funan were no imports—they appear to have been locally produced from Roman coins circulating in the area.

It is beyond fascinating: ancient historians have frequently emphasized the potential of Roman coins for propagandistic messagery within the empire, and this desire to communicate clearly articulated images of power transformed coin design in the first centuries of our era. Yet these pendants show how such imagery not only could but also did transgress cultural boundaries, and found new use in a world further away from the Roman emperor than many Romans may have been able to imagine.

B. Borell (2014), ‘The Power of Images – Coin Portraits of Roman Emperors on Jewellery Pendants in Early Southeast Asia’, Zeitschrift für Archäologie Aussereuropäischer Kulturen 6: 7-44.

Urban images (1) – Damnatio memoriae at Roman Corinth

Miko Flohr, 07/02/2020

Corinth, Roman Forum. Remains of an inscription showing the partially erased names and titulature of a Roman Emperor. The text is thought to have read:

Imp(erator) Caesar divi M(arci) Antonini Pii Ger[m(anici) f(ilius) divi Pii n(epos) divi Hadriani pron(epos) divi Traiani Parthici abn(epos)] / divi Nervae adnepos [[M(arcus) Aurel(ius) Comm[odus 3]]] / ex testamento Cornel(iae) Baebiae fecit cur[avitque 3] (AE 1947, 90)

The emperor whose main name is removed is thought to have been Commodus, who was murdered in 192 CE, triggering a major succession crisis that eventually would bring to power Septimius Severus. What is interesting, here, is that only part of the name is erased. The text now essentially reads ‘That guy, the son of Marcus, grandson of Pius, great grandson of Hadrian, great great grandson of Trajan, great great great grandson of Nerva, whose name shall go unmentioned, …’

Global Romans (5): The archaeology of Ptolemy’s ‘Golden Peninsula’

Miko Flohr, 04/02/2020

Over the last weeks, this series has focused on Java and Bali, and it became clear that ties between the Far East and the Far West in the first centuries of our era were closer and more direct than one would expect on the basis of handbooks and scholarly discourse – even though it is hard to pinpoint to unequivocal evidence for direct contact. The world beyond India – to look at it from a Greco-Roman perspective – was of course much bigger than the Indonesian archipelago, and, indeed, it is clear that Hellenistic and Roman authors knew that the world did not end at India. But what, exactly, they knew is much less straightforward, as everything they write is either extremely sketchy or completely unreliable (in most cases, it is actually both).

Ptolemy’s Geographica and the world beyond India

Particularly tantalizing is the work of the mid-second century CE scholar Ptolemy, whose Geographica lists the coordinates of not fewer than 8000 places in the known world that together form the basis of a map covering the area from the Atlantic in the Far West to far beyond the Indian subcontinent in the Far East. As part of his introduction to the map, Ptolemy gives a detailed description of a sea voyage between the southernmost tip of Sri Lanka, and a mysterious place named ‘Kattigara’. This settlement, of which very little else is known, was to be found beyond a place referred to as the ‘Golden peninsula’ (Ptolem. Geogr. 1.13-14). Ptolemy based his account on earlier Greco-Roman authors, some of whom may have visited the area personally, though we know very little about their details. A major source was the early second century CE geographer Marinos of Tyre.

A Byzantine Greek world map according to Ptolemy’s first (conic) projection. From Codex Vaticanus Urbinas Graecus 82, Constantinople c. 1300.

So: where are the places where Ptolemy is talking about? His work gives the coördinates, but these are generally not thought to have been fully reliable for the far East. As far as the ‘Golden peninsula’ is concerned, it is interesting that Ptolemy actually states the peninsula needed to be crossed rather than circumnavigated. For this reason, people have thought, probably rightly, that the ‘Golden Peninsula’ in fact is the Thai-Malay peninsula. If the Golden peninsula is in modern-day Thailand and Malaysia, Kattigara probably is to be sought somewhere in Vietnam, but this is something for a later moment.

Archaeological Excavations at the Isthmus of Kra

The question, then, becomes whether this greater Greco-Roman knowledge of the Thai-Malay peninsula also translates to a greater archaeological visibility of cultural contact. Has more Roman stuff been found in the ‘Golden peninsula’ than was the case at Java? A recent overview of Greco-Roman finds from the upper part of the Thai-Malay peninsula suggests that, in fact, this was the case. Particularly remarkable are the finds from a number of sites in the region of the so-called Isthmus of Kra in the environs of modern Ranong. This is the point where the Thai-Malay peninsula is at its narrowest, and it appears to have been a major crossing point in the first centuries of our era – it may have been the point to which Ptolemy is referring, but this is not certain.

Roman cameo fragment from the Augustan Period found at Bang Kluai Nok, Thailand

Archaeological excavations in the region of the Isthmus of Kra have turned up large numbers of objects from Han-China together with certain amounts of artefacts from the Roman world. Compared to what we have been seeing at Java, the identification of these artefacts as ‘Roman’ is much more straightforward, and they were larger, and more luxurious, including finely decorated first century AD glass vessels, several finely decorated gemstones, and even a cameo fragment from the Augustan period with a clearly recognizable Satyr – Greco-Roman iconography had, to some extent, entered Thai material culture.

Does this mean that we have to reckon with substantial quantities of Greco-Roman travellers all the way to the environs of Ranong in Thailand? That would be stretching the evidence beyond the point of credibility. But, if we combine the textual sources with the archaeology, the difference between the Thai-Malay peninsula and the Indonesian archipelago is certainly telling, and it suggests that even if it was quite an exotic destination for someone from the Roman Mediterranean, the Thai-Malay peninsula at least could be a destination.


B. Borell, B. Bellina & B. Chaisuwan (2014). ‘Contacts between the Upper Thai-Malay Peninsula and the Mediterranean World’ in: Before Siam was Born: New Insights on the Art and Archaeology of Pre-Modern Thailand and its Neighbouring Regions, 98-117.

J.L. Berggren & A. Jones (2000) Ptolemy’s Geography: An Annotated Translation of the Theoretical Chapters. Princeton University Press.

Global Romans (4) – Late Roman coins in East Java

Miko Flohr, 28/01/2020

The problem with beads, of course, is that they are beads. They’re small, unremarkable objects that do not carry a very explicit intrinsic message. Of course, they can mean the world for the individuals to which they belong, as part of a necklace or bracelet, but it is very hard to see how the people wearing these objects could know, and give meaning to the provenance of these beads. Indeed, the small quantities of ‘Roman’ glass beads found on Bali make it likely that they were not aware at all of their exotic origins. Just as Pliny did not know about the Maluku islands, people in the Indonesian archipelago may not generally have known that something like the Roman Empire existed – indeed, many may barely have been aware of the existence of India or China.

Bronze coin of Constans (347-348 CE). Source: wikimedia.

Still, there were of course classes of objects that revealed much more direct links to the Roman world, and coins are the most durable example. Indeed, Roman coins are known to have travelled significant distances, and they carried explicit imagery, including, from the Augustan period onwards, the portrait of the emperor under whose authority coins were struck. Even without any contextual information, these coins offered a direct visual link to their place of origin, and even people who would not know anything about the Roman empire (or, indeed, about coinage), would be able to recognize that this was an object with visual information that came from some other place – thus creating at least the notion that that other place existed – whatever it was – and offering a starting point for a narrative about that place.

The question is whether Roman coins travelled all the way to the Far East, and, if so, how. There is ample evidence for Roman coins in India from the Augustan period onwards – I will undoubtedly write about this at a later moment – but beyond the Bay of Bengal, things were more complicated. Nevertheless, perhaps surprisingly, there are reports of Roman coin finds from South-East Asia, even if they often lack detailed information about the life course and ‘taphonomy’ of these coins—how they travelled, got lost, and ended up buried. A recent (2018) article by a Hungarian team has brought together and critically assessed all the evidence known so far.

On Java, one discovery stands out. In 2017, a substantial set of ancient coins was reportedly found in the Brantas river, near Trowulan – the former capital of the (medieval) Majapahit empire that I briefly mentioned two weeks ago. The finds, made by local gold panners, included ten Roman coins, and a number of medieval coins from China. What is interesting, however, is that the Roman coins all date from roughly the same era – seven were struck between the 320s and the 360s CE, two date to the 270s, one to the 220s – if found in a proper archaeological context, they would be interpreted as reflecting a live set of coins from the later fourth century CE. Interestingly, the coins represented little monetary value: they were all bronzes, and belonged to the smaller denominations.

Siliqua of Julian II (348-364 CE). Source: wikimedia.

What is going on is very hard to tell: even if the Roman coins together kind of make up a credible wallet from the later fourth century, it is unsure when and how they travelled, why they ended up in Eastern Java, and how they ended up in the Brantas river together with several much later Chinese coins. The fact that they are bronze coins, and not silver or gold makes it unlikely that they represent a modern collection that was simply dumped in the river, though the option cannot be entirely discarded. That there are few coins from after the fourth century suggests that these coins originally travelled together, and did so during Late Antiquity, but they can barely have been brought in as a means of payment.

Two things should be emphasized. First, it appears to be an isolated find without a context. Loose coins appear to be found elsewhere in Indonesia – at least, they appear on the market – but not (yet) during excavations, and not in sets. Second, these coins constitute no evidence for a widespread circulation of coins that means that people actually saw these objects on any regular basis. Future finds could change that, but as things stand, the Roman coins from Trowulan are (and should be seen as) little more than a historical anecdote.

Still, it can be argued that, given the small monetary value of the ensemble, one option is that this coin find reflects a Late Roman individual traveling all the way to the far east. Another possibility is that these coins reflect an exchange of gifts between a Late Roman traveller and someone from East Java, which may also have taken place in India. There are multiple other possible scenarios, and we should be careful not to fall into the trap of overinterpretation. Nevertheless, the find is tantalizing, and begs for more.

Hoppal, K. et al. ‘All that glitters is not Roman’ Roman coins discovered in East Java, Indonesia. A study on new data with an overview on other coins discovered beyond India,’ DissArch 3.6 (2018): 461–492. DOI: 10.17204/dissarch.2018.461

Middenschooldebat vraagt om veel meer dan alleen feiten uit de onderwijssociologie

Miko Flohr,

In een opiniestuk in de Volkskrant betoogden Louise Elffers en Thijs Bol afgelopen maandag dat vroeg selecterende onderwijsstelsels—zoals het Nederlandse—hun ‘uiterste houdbaarheidsdatum’ hebben overschreden en dat het tijd is om ook in Nederland werk te maken van latere selectie in het onderwijs. Hun betoog volgt op de discussie die zich vorige week ontspon nadat een aantal onderwijsorganisaties had geadviseerd om de leeftijd waarop leerlingen definitief kiezen voor het niveau waarop ze onderwijs gaan volgen te verhogen van twaalf naar vijftien. Elffers en Bol roepen op om het debat op basis van feiten te voeren, en dat lijkt me inderdaad een uitstekende zaak.

Oog voor mitsen en maren

Het is daarbij echter wel belangrijk om ook oog te hebben voor de mitsen en maren die er aan zo’n plan zitten, en de manier waarop Elffers en Bol stellen dat de tegenargumenten tegen de invoering van latere selectie louter berusten op onjuistheden en misverstanden doet vermoeden dat er op dat punt nog wel wat ruimte is voor enige tegendruk. Dit des te meer omdat de ‘feiten’ die ze zo overzichtelijk op een rijtje zetten ook nog wel het een en ander onbesproken laten.

Voor de duidelijkheid, vooraf: nee, de schrijver dezes is niet van mening dat alles nu rozengeur en maneschijn is, en ja, hij onderschrijft het idee dat onderwijs de sleutel is voor allerlei vormen van sociale mobiliteit en hij weet dat er op dat punt in Nederland veel beter kan, en moet. Het is daarnaast helder dat het voor allerlei groepen leerlingen beter is om vooral niet te vroeg in een bepaald leertraject geplaatst te worden. Het lijkt me goed op dit punt het wetenschappelijke discours tot ons te nemen en te omarmen, en het stuk van Elffers en Bol levert daarvoor de nodige aanknopingspunten. Dat is een goede zaak.

Tegelijkertijd is er een gapend gat tussen wetenschappelijk onderwijssociologisch discours enerzijds en, anderzijds, de rauwe realiteit van grootschalige onderwijsvernieuwing. Precies op dit punt zetten Elffers en Bol hun lezers op het verkeerde been. Het zal allemaal wel meevallen, suggereren ze, want er gebeurt al heel veel binnen het huidige stelsel, en het is deze keer niet iets dat van bovenaf wordt opgelegd, maar het komt vanuit de sector zelf. Maar: wat Elffers en Bol propageren is een zeer fundamentele stelselherziening met zeer verstrekkende gevolgen voor de dagelijkse lespraktijk waarvan het maar de vraag is of grote delen van het onderwijsveld er om zitten te springen.

Verstrekkende gevolgen

Het is ronduit onbegrijpelijk dat ze de impact van het plan zo bagatelliseren: de eerste drie jaar van de middelbare school gaan er totaal anders uit zien, met klassen waarin meerdere groepen leerlingen tegelijkertijd op verschillende niveaus werken aan stof over een vergelijkbare thematiek, en waarin docenten een heel andere rol spelen dan nu het geval is. Het is mooi dat er meerdere scholen zijn waar dit al gebeurt, en het is prachtig dat het daar goed werkt, maar het vergt een totaal andere manier van werken voor alle docenten op alle scholen in Nederland. Hoe groot is het percentage van hen dat daar, onder het huidige belonings- en werkdrukregime, warm van wordt? Daarover horen we eigenlijk niets.

Daarnaast verzwijgen Elffers en Bol dat een structurele ophoging van de selectieleeftijd naar 15 jaar de facto betekent dat de zelfstandige gymnasia in Nederland de deuren zullen moeten sluiten. Hoe bezopen dat ook moge klinken, dat kan, uiteraard, in principe, een beleidskeuze zijn. Het is alleen wel een keuze met enorme consequenties, en een keuze die je niet zomaar stilzwijgend kan maken. Als Elffers en Bol vinden dat de zelfstandige gymnasia in Nederland maar opgeheven moeten worden, dan moeten ze daar niet over zwijgen, maar het expliciet in hun stuk zetten en onderdeel maken van hun argument, en niet pretenderen dat zo’n verandering niet ingrijpend is. Dat is misleidend.

Tot slot een ander, misschien nog wel fundamenteler, punt. Het onderwijskundig onderzoek dat de effecten van stelselherzieningen meet, en dat een belangrijke pijler is onder het idee dat zo’n stelselwijziging gunstige effecten heeft, is gebaseerd op het effect van stelselwijzigingen uit de jaren zestig en zeventig van de twintigste eeuw. Met die analyse is weinig mis, maar de vraag of zo’n benadering ook voorspellende waarde heeft voor stelselveranderingen die meer dan een halve eeuw later op de rol staan is geen onderdeel van het onderwijskundige discours. De wereld van 2020 is allang niet meer de wereld van 1965, en als je in 2020 aan een knopje draait leidt niet noodzakelijkerwijs tot datzelfde gunstige feedbackeffect dat in 1965 optrad. Pretenderen dat dat anders is, is het gebruiken van een in wezen historische dataset voor een universalistisch argument, en dat lijkt me, zacht uitgedrukt, erg onvoorzichtig.

Feiten zijn feiten, maar zijn het ook alle feiten?

Feiten zijn feiten, en we moeten dit debat zeker op basis van feiten voeren, maar het onderwijswetenschappelijke discours levert bij lange na niet alle feiten die we nodig hebben om een goede beslissing te kunnen nemen over een verregaande stelselwijziging. Dat is volstrekt geen diskwalificatie van het goede, en interessante werk dat wetenschappers in deze disciplines verrichten. Het is wel een realitycheck: we hebben een complex onderwijsstelsel dat kraakt en piept van onderfinanciering, en waar hoogopgeleide vakmensen zich tegen (te) weinig salaris (hard) uit de naad werken om hun (veel te) grote groepen leerlingen richting de respectievelijke eindtermen te dirigeren. Het is leuk dat onderwijssociologisch onderzoek impliceert dat er van alles anders moet, maar zonder een doordacht veranderplan dat geworteld is in deze realiteit is iedere discussie simpelweg luchtfietserij. En nee, een einddoel is geen veranderplan. Het heeft geen zin te pogen een rivier over te zwemmen zonder dat je weet hoe je aan de overkant moet komen, hoe mooi die pot met goud er ook staat te glanzen. Soms kun je, alles overwegende, beter besluiten de sprong toch maar niet te wagen. Bovendien: het onderwijskundige discours mag dan suggereren dat het aan de overkant beter is, maar hoeveel beter het precies kan of zal zijn is vooralsnog veel minder helder. Niet alles wat blinkt is goud.

Er zijn ongetwijfeld nog meer haken en ogen te noemen. Moet dan alles maar bij het oude blijven? Nee joh. Ik ben er een groot voorstander van dat meer scholen, op basis van onderwijssociologische inzichten besluiten selectiemomenten uit te stellen, of gefaseerd gaan selecteren. Dat kan je faciliteren en stimuleren door positieve financiële prikkels, en door drempels weg te nemen, en kan al heel veel doorstroomproblemen verzachten – zonder alles meteen dwingend, stelselbreed, op te leggen. Ik ben er ook erg voor om de zelfstandige gymnasia te stimuleren om bij het begin van 4VWO een regulier tweede instroommoment te creëren, zodat zij minder losstaan van de rest van het onderwijssysteem (want dat is inderdaad wel een punt van zorg). Daarnaast moet het normaal worden en blijven om op- en af te stromen. Ook dat kun je faciliteren, met relatief simpele maatregelen, en met relatief weinig geld. Als zo, van onderop, over vijftien jaar, een realiteit is ontstaan waarin ons onderwijs binnen het huidige stelsel op de geobserveerde knelpunten veel beter functioneert dan nu het geval is, kunnen we altijd nog eens overwegen of de keuzeleeftijd niet toch generiek en van overheidswege naar 15 jaar kan. Tot die tijd passen ons niet alleen de ‘feiten’, maar ook een gezonde dosis realisme omtrent de complexiteit van het stelsel dat we hebben, en omtrent de huidige staat van de onderwijssector.

Global Romans (3) – A Roman Bead from Bali

Miko Flohr, 21/01/2020

If the evidence from early first millennium CE West Java discussed last week leaves no doubt that parts of the island were integrated into the larger regional maritime networks of the era, there was no unequivocal evidence of stuff from beyond India arriving in Java – there is a clear possibility that some of the artifacts discovered at Batujaya came from the Far West, but no direct proof. Yet over the course of the last two decades, more archaeological work has been done elsewhere in Indonesia, and most of these projects have equally been interested in the long-distance contacts between the Indonesian archipelago, Asia, and the rest of the ‘old’ world.

An interesting case is that of Sembiran and Pacung on Bali. Bali, of course, lies immediately east of Java, but in the larger scheme of things that means that it is further away from the main trans-Asian seafaring routes, and therefore, its integration in ‘global’ networks of exchange was a bit more complicated. To put it a bit more bluntly: while North-West Java is relatively well-connected to a variety of places simply because of its location, Bali had mostly one claim to fame, and that is that it was (roughly) en route to the Maluku Islands, where unique spices could be obtained, and one of the debates amongst scholars is how intensive this spice trade had become in the early first Millennium CE.

Cloves from the Maluku archipelago

Some trade clearly was already going on: both in Han China, and in the Roman empire, people knew of the existence of cloves, which only grow in the Maluku Archipelago. Indeed, the first century CE Roman author Pliny is commonly thought to refer to the clove in his encyclopedia of nature, calling it a ‘grain resembling that of pepper, but larger and more brittle, which was imported to the Roman world ‘for the sake of its scent’ (Natural History, 12.30). Pliny, however, reported hearsay that the clove came from the ‘Indian lotus tree’ rather than from the Indonesian archipelago – Greco-Roman topographical knowledge beyond the Bay of Bengal was extremely sketchy.

Yet if products from the Maluku Islands could make it to the Greco-Roman world, then the reverse is also true, and Greco-Roman stuff might easily have ended up on the Maluku Islands, or in places along the way – such as Bali. The excavations in Sembiran and Pacung served to find out how coastal communities in Northern Bali interacted with the networks of trade that made this possible, and the excavators have analyzed a number of finds – grave goods – to understand these connections. Obviously, these included ceramics. As happened at Batujaya, there was a clear presence of Indian rouletted ware in layers that could be dated to around the start of the Christian era, but it also seems as if some Indian-looking pots were actually local imitations (suggesting that people liked this stuff, and wanted to replicate it).

Roman Glass bead (not the one from Bali)

More interestingly, however, the excavators also chemically analysed a whole set of glass beads – there were different traditions of glass-making in the ancient world, and the chemical components of glass can be used to understand, in very general terms, where it is coming from. Most beads came from mainland southeast Asia, several could be associated with India, but three beads, of which one from a firmly dated (early) context, were made of soda Natron glass, a tradition firmly (and only) associated with the Roman world – indeed, the natron used in the Roman world all came from Wadi-el-Natrun in Egypt. So here, one could say, we have a concrete and undisputable example of Roman stuff in the Far East from what in Greco-Roman terms counts as the ‘Early Imperial period’. Three beads, each of barely half a cm in size. It is not much. But it is definitely something.

Calo, A. et al. (2015) ‘Sembiran and Pacung on the north coast of Bali: a strategic crossroads for early trans-Asiatic exchange’ Antiquity 89: 378-396. doi: 10.15184/aqy.2014.45

Global Romans (2). Batujaya and the global connections of Java in the first century CE

Miko Flohr, 14/01/2020

So, let’s just start with the place that got me into this thing in the first place. What was happening on the island where my grandparents were born when the Romans were building their riverine frontier in The Netherlands? Very little of the pre-colonial history of Indonesia is part of the western historical canon. Some people will know that before Islam became the dominant religion on Java – which happened only in the 15th century – the island was under Hindu-Buddhist influence. A few, perhaps, will know that there once was an empire that ruled over larger parts of the Indonesian archipelago from its base in Trowulan, East Java (incidentally, that is within walking distance of the place where in 1920 my grandmother would be born). This so-called Majapahit Empire had its golden age in the 14th century and it was the last in a series of Hindu-Buddhist kingdoms that had ruled over parts of Java from the fourth century CE onwards.

Java before Indianization

In the first century CE, all of this was in the far future. ‘Indianization’ (as it is called by some) had not yet truly begun, and there is no evidence that the Indian religions had established a meaningful presence in Indonesia. In fact, no historical sources from the period exist: the oldest texts on Java date to the reign of an early fifth century (CE) king of the kingdom of Tarumanagara, who was called Purnawarman. They are written in Sanskrit using the so-called Pallava-script – indeed, both the language and the script came from India.

Buni culture pottery, coastal West Java c. 400 BCE to 100 CE . Collection of National Museum of Indonesia, Jakarta.

To understand what is happening on Java before this period, one therefore has to rely on material culture. For a long time, this quite simply meant that, basically, nothing was known about what was happening, and still, the number of seriously investigated archaeological sites with evidence from the last centuries BCE or the first centuries CE on Java remains limited.

This is not to say that nothing is there. A couple of archaeological sites in North-West Java, have yielded remains from datable contexts that give some clues about some of the developments of the period. The evidence remains in many ways elusive: no built structures are known, and most finds are artefacts coming from funerary context.

Buni and Batujaya

The local pottery tradition of the time has become known as the ‘Buni culture’, after one of the principal findspots. It is characterized by very simple, incised geometric decoration in a style that somewhat resembles contemporary pottery traditions elsewhere in South-East Asia. Even if all of this pottery was locally made, this suggests that the north coast of Java had already been integrated in larger regional networks of maritime (cultural) exchange.

Batujaya, West Java: one of the excavated temples amidst rice fields

One of the best published sites with evidence from the period is Batujaya, not far from Buni. This site, discovered only late in the 20th century, is best known for its ‘medieval’ (fifth to ninth century CE) Buddhist temple complex, of which ample remains are visible on-site, but a substantial number of burials from the first two centuries CE have been found in lower levels (Manguin & Indradjaja 2011).

The grave goods from these burials give a good indication of the material culture of North-West Java, which was a mixture of locally produced objects and, increasingly, imports. As in multiple other sites from the period, grave goods included so-called ‘rouletted’ ware and stamped ware from India, but also a gold object that the excavators have associated with ‘Persia’ (or, in this period, rather, the Parthian empire).

Textiles from the Far West?

Still, there may be a bit more. In a 2015 article, the excavators of Batujaya presented a laboratory analysis of some very fragmentary textile remains found in three of the tombs, arguing that one of the texiles was, in fact, asbestos, and suggesting that it might come from what they call the ‘Roman Orient’ (Cameron et al. 2015). This is certainly a possibility – if China in this period were getting their asbestos (indirectly) from the Far West, as their sources suggest, the same might be true for North-West Java. However, other options (particularly India) do not seem to be off the table, and the textile could not be securely provenanced by any technological means.

Nevertheless, whatever was happening on Java, their coastal communities were well-connected with the wider world, and – more interestingly – they had direct access to a network that also had connections to the Roman Empire of the first century CE As ever, the clearest indication for this comes from ceramics: the very same Indian ceramics that have been found in Batujaya has been found, in certain quantities, in the Roman harbours of Berenike and Myos Hormos on the Red Sea in what is now Egypt (Tomber 2000). Maybe, there were, in the first century CE, fewer handshakes between the Limes in Germania Inferior and Java than we would perhaps be inclined to imagine.


Cameron, J., A. Indradjaja, P.-Y. Manguin (2015), ‘Asbestos textiles from Batujaya (West Java, Indonesia): Further evidence for early long-distance interaction between the Roman Orient, Southern Asia and island Southeast Asia’ Bulletin de l’École française d’Extrême-Orient 101:159-176.

Manguin, P.-Y., Indradjaja, A. (2011) ‘The Batujaya Site: New Evidence of Early Indian Influence in West Java’ in: P.-Y. Manguin, A. Mani, and G. Wade (eds) Early Interactions between South and Southeast Asia. Reflections on Cross-Cultural Exchange. Singapore: ISEAS, 113-136.

Tomber, R. (2001) ‘Indo-Roman trade: the ceramic evidence from Egypt’ Antiquity 74: 624-631.

Global Romans (1). The classical world through post-colonial eyes

Miko Flohr, 07/01/2020

I have lived inside the boundaries of what used to be the Roman empire for my entire life. Born in Rotterdam in The Netherlands, I grew up in Brabant, in the south of the country, and went to the university in Nijmegen, where they were still excavating the legionary camp as I arrived in 1995 to study Classics; every year or so, newspapers would report new local discoveries – burials, structures, or roads. Later, in Oxford, we lived within a couple of hundreds of meters of the old Roman road between Silchester and Towchester. Since moving back to the Netherlands, we live right between the Roman castellum at Valkenburg, and the small town of Forum Hadriani at Voorburg, and I happen to work within fifty meters of the northern boundary of the Empire – when I go on my bike from home to work, I often cross it twice. I know no better than that the object of my study is a regular part of the past of the places that I have called home.

Birth places of my ancestors, five generations back. Top: The Netherlands. Bottom: Java, Indonesia.

Oddly enough, it took me some time to realize that this does not also mean that my deeper personal roots lie within the empire. Actually, as the map above shows, the reverse is true. My mother was born just north of the Rhine, in Arnhem, her parents in Arnhem and Groningen. Part of my grandfather’s family can be traced back to the riverine area that was once just inside the empire (but only just); however, my grandmother’s ancestry almost entirely came from Frisia, which never was part of the Roman realm. More radically, my father was born on the other end of the world, in Southern Sumatra, in what then still was called the Dutch East Indies by some. Both of my grandparents who were born on Java and had mixed Javanese-European ancestry. Only in 1954, they came to Europe. It is safe to say that there is not an awful lot of Greco-Roman blood running through my veins.

In a way, of course, this is pompous nonsense. Family histories can only be traced back to the early modern period, and we are shaped by them only to a fairly limited extent even if they are of great antiquarian personal interest. Also, there is no intrinsic meaning attached to living within or outside the former boundaries of an empire that has been gone for a millennium and a half apart from the fact that Roman stuff may be found in your back yard or around the corner. Antiquity is dead and gone, and no-one in this world, except for the pope maybe in Rome, entertains a living relation with the world of the first centuries of our era.

Still, place matters, and it is kind of ironical how we, in the Netherlands, and in larger parts of North-Western Europe (never mind North-America), have somehow decided, long ago, that our collective historical narratives start from Greece and Rome, far away in the Mediterranean. In a way, we have artificially appropriated a past that is wholly or partially unrelated to the land where our families come from and we have made it our own. Additionally, we have asked people arriving in our country from elsewhere (like my father’s family) to buy into this narrative even if it essentially excludes the places where they (also) come from. It is surely a tempting narrative – I am certainly not the only European with roots in the Far East to study Greco-Roman ‘classics’. But it is also a bit odd, if you think about it – it is strange to spend your professional life within a historical tradition that has nothing to say about the land where part of your family comes from.

Personally, I only gradually came to be bothered by the geographical boundaries of ‘our’ ancient world. Of course, I learned about Rome’s intensive trade with India in the first and second centuries CE, and I read stuff about China under the Qin and Han emperors, but the actual position of Rome, and its empire, in the wider world of the early first millennium CE is something that only came onto my radar over the course of the last two years or so; perhaps, this is due to the increasingly heated debates about the extent to which European perceptions of the past are still firmly rooted in ideas defined in the era of European colonial empires. Should not a post-colonial view on the history of the world also have implications for the way in which we approach the Ancient Mediterranean, and if so, which?

I would argue that it is the task of Ancient Historians to begin to set this record straight. Some of us, of course, have already begun spinning historical threads connecting the Greco-Roman world to the world around the ancient Mediterranean, including India and China. Yet it is safe to say that very little of that has really entered the central debates in our field, or reached a broader public. Indeed, if Greco-Roman scholars experiment with the concept of globalisation, they mostly use it as a concept within the confines of the classical world – that’s how provincial we (still) are! Indeed, the wider world is only rarely part of what we teach – our world, too often, still stops where (and when) they stop writing Greek and Latin. There are some prominent exceptions, but it also seems that many of us simply don’t know very well what the hell is happening out there. This needs to change – we need to change.

So here we go. What I will be doing is write a number of blogs on the ‘echoes’ of Rome in the world outside the boundaries of the empire – and vice-versa. One could argue that this still is, essentially, a Romano-centric approach, and that certainly is true. But it also offers a powerful perspective, and what I hope is that it does give me some clues in connecting all the dots – after all, I have a direct personal interest in doing this. I have no fixed schedule, and no clear timetable for this, and many other obligations, so we’ll see how it goes. This might be a five-year project, on and off, but eventually, we’ll get there, and it will help me understand how the Greco-Roman world in our handbooks related to and interacted with the world beyond the Rhine, the Danube, and the Syrian, Arab and North-African deserts. Maybe, it will help some of you as well.

Note: The title of this series is resembling that of Frits Naerebout’s recommendable article ‘Global Romans? Is globalisation a concept that is going to help us understand the Roman empire?’ (Talanta 38-39 [2008], 149-170), which can be found here. It deserves some emphasis, however, that it lacks the question mark.