I have recently come up with a new picture of premodern Southeast Asian history by revisiting Chinese, Siamese and Cambodian sources.
In particular, I have come to conclude that there was never a place called “Srivijaya,” a kingdom that was supposedly “rediscovered” by George Cœdès in 1918. Instead, I am convinced that the information in Chinese sources that was used to create a history for “Srivijaya” was actually about “Cambodia.”
In addition to information in Chinese sources, the scholars who developed the idea that there had once been a kingdom called “Srivijaya” also employed sources written in languages like Persian and Arabic.
I have recently started to look at those materials (in translation) and have discovered that there are just as many problems with the ways that people have interpreted those sources as there are with the ways that scholars have interpreted Chinese sources.
There is apparently an important book entitled A Study of the Arabic Texts Containing Material on South‑East Asia by G. R. Tibbetts. Published in 1979, it is out of print and there doesn’t appear to be a digitized version out in cyberspace anywhere.
So, I have no access to that book.
What I have been looking at is the stuff that was produced in earlier periods. In the nineteenth century, for instance, there were various French authors who translated Arabic texts into French. Then in the early twentieth centuries there were scholars who compiled together extracts from these translations and commented on them.
Examples of this are Colonel G. E. Gerini’s 1909 Researches on Ptolemy’s Geography of Eastern Asia (Further India and Indo-Malay Archipelago) and Gabriel Ferrand’s 1913 Relations de voyages et textes géographiques arabes, persans et turks relatifs à l’Extrême-Orient du VIIIe au XVIIIe siècles (traduits, revus et annotés par Gabriel Ferrand).
All these works are frustrating to read and difficult to use. First, one should never trust a nineteenth-century translation. Such works were created in a different age and need to be re-evaluated, but I don’t have the ability to do that with texts in Arabic.
Second, instead of just providing information that is in Arabic texts, scholars like Gerini made decisions about what information was or was not accurate in texts and then presented the information in these texts in light of these perceptions.
Finally, after the idea that there was a kingdom called “Srivijaya” emerged, there were scholars who read (the translations of) Arabic sources with that idea in their minds.
As a result, the translations of Arabic texts and the discussions of these works have produced many conflicting ideas that scholars have not been able to resolve. However, I think that when we revisit these sources after realizing that there was no “Srivijaya” and after seeing what the Chinese sources that were used to create the history of “Srivijaya” actually refer to, we gain an opportunity to reinterpret Arabic texts about Southeast Asia and to arrive at a much better understanding of what those texts record.
One issue that has confused scholars, but which the Chinese sources that have been used to write about “Srivijaya” make clear, is the fact that there historically were two “Javas” in Southeast Asia.
There is an island today in Indonesia called “Java” and that island is clearly recorded in Chinese sources starting in the thirteenth century as “Zhuawa” 爪哇.
Prior to that point, there was a place recorded in Chinese sources that had a similar sounding name. It was recorded in Chinese as 闍婆, which can be rendered today as “Shepo” or “Dupo.” That may not sound like “Java,” but in the first millennium AD, those characters were used to transcribe Sanskrit terms, and when they were used to do that, 闍 could represent the sound “ja” and 婆 could represent the sound “ba” or “pa” or “va,” etc.
Therefore, what we now pronounce as “Shepo” could have been pronounced in the past as “Jaba” or “Japa” or “Java,” etc.
Was Shepo and Zhuawa the same place? Chinese historians who wrote around the seventeenth century claimed that they were. They said that the Zhuawa (Java island) of their time had previously been known as Shepo.
In the article that I have written on “Srivijaya” (it should have been published on 1 June 2022, but we’re still waiting. . .), I demonstrate that this was not the case. Instead, I argue that Shepo was located on the Malay Peninsula in the area of what is today Songkhla and Patthalung provinces in Thailand.
How do I demonstrate this? First, the geographical descriptions of Shepo in Chinese sources describe a place from where the sea can be reached in four directions, and that describes the narrow area of the Malay Peninsula where Songkhla and Phatthalung are located much better than the large island of Java.
Further, the fact that information is recorded about reaching the sea in different directions indicates that people were actually moving across the land to the sea, and the logical reason to do so would be to trade. The Songkhla/Phatthalung region is the perfect place for trans-peninsular trade and there is plenty of archaeological evidence that such trade took place in the past.
Second, there is some information in Chinese sources about the geography of Shepo which makes it clear that there was something unique about the sea to the east of Shepo, and the area of Songkhla/Phatthalung is unique in that there is a large lake that is separated from the sea by a narrow piece of land known now as Sathing Phra.
What I argue in my article is that Shepo, a place on the Malay Peninsula, was a very important trading center until the Siamese kingdom of Ayutthaya emerged in the middle of the fourteenth century and brought that region under its rule, thereby disrupting the trans-peninsular trade network and bringing an end to Shepo as an autonomous kingdom.
Meanwhile, starting a bit earlier, in the late thirteenth century, a kingdom on the island of Java, Majapahit, became power and expanded its influence through the region. As it did so, this other “Java” became more widely known.
In the case of Chinese sources, there is a brief period in the second half of the fourteenth century when there are records about both Shepo and Java. Then Shepo disappears from the sources (as, I argue, it came under the control of Ayutthaya).
So, what do Arabic sources show us? Having spent the past couple of days looking around in them, I can easily see that they tell the same story as the one I have just told above.
There is a place that gets mentioned frequently in Arabic sources that is clearly Shepo. It gets rendered in many different ways, Zabag, Zabej, Djawaga, however, these variant names all clearly indicate the same place, and the information about that place fits with the geographic location of Songkhla and Phatthalung.
In particular, there were Arab authors who noted that a distinct feature of Zabag/Zabej/Djawaga was that it was located at a place were the sea and fresh water interacted in a kind of bay area.
For instance, Gerini cites an author by the name of Abu Zaid (880-916) who recorded the following: “The palace of the Maharaja fronts a thalaj, or estuary (marsh, lagoon), formed by an inlet of the sea. This is invaded by sea-water at flow-tide, but the water therein turns out fresh at the ebb.” (560)
Gerini has a long footnote in which he discusses this term “thalaj.” He is not clear where it comes from and therefore can’t be certain of what exact type of body of water it refers to. I also do not know, however, the inland body of water at Songkhla/Patthalung where sea water and fresh water meet is called a “thalei” now in Thai. I do not know the etymology of that word.
That said, the environment that Abu Zaid described, and the term he used, both strongly suggest the area of Songkhla/Patthalung and Sathing Phra.
An author by the name of Mas’udi made a similar comment in 943: “The throne of the king of Zabej overlooked. . . a small marsh which communicated with the principle bay (estuary of the river) of Zabej. The flow let the sea-water into that bay, and the ebb allowed the fresh water to freely run down. (561)
Zabag/Zabej/Djawaga thus seems to point to a place like Songkhla/Patthalung and Sathing Phra, as does the Shepo in Chinese sources. When we then remember that Shepo was probably pronounced as something like “Jaba” or “Java” then it is easy to see that the Arabic terms approximate this pronunciation.
Further, just as there was a place by a similar-sounding name that came to replace Shepo in Chinese sources around the fourteenth century, I think we can detect a similar phenomenon in Arabic sources.
In the early fourteenth century, for instance, an author by the name of Nowairi, provided information about different “seas” in Southeast Asia. Zabag/Zabej/Djawaga was in one sea and a place called Djawa was in another. (Ferrand, 395)
At roughly this same time in Chinese sources, we can also see the simultaneous existence of Shepo and Zhuawa, whereas prior to that point, there is only information about Shepo, and after this point, there is only information about Zhuawa.
Where things get confusing, and where future research will require a lot of work, is that Arab sources indicate that the ruler of Zabag/Zabej/Djawaga ruled over many different “islands.” It seems clear that some of these places were not actual islands, but locations on land masses, like Kedah on the Malay Peninsula.
Scholars have worked hard to try to identify where these places are, however, I would argue that they have been looking in the wrong direction. Gerini and many subsequent scholars have tried to place Zabag/Zabej/Djawaga on the island of Sumatra, and some have even tried to equate it with “Srivijaya.”
Meanwhile, although Arabic sources repeatedly indicate the importance of Zabag/Zabej/Djawaga and say that other places were under its authority, geographer Paul Wheatley inexplicably ignored it in his 1961 The Golden Khersonese: Studiese in the Historical Geography of the Malay Peninsula before A.D. 1500. He also only made a passing reference to Shepo in that work.
So, for over a century, most scholars have been imagining that Arabic sources contain information primarily about Sumatra, Java and some parts of the Malay Peninsula, like Kedah. However, I don’t think that is correct.
What I see at the moment is the following: The most important trading place in the region that we see in Arabic sources is Zabag/Zabej/Djawaga. This place is the same place as Shepo and was located at Songkhla/Phatthalung/Sathing Phra.
What made this place so important was the fact that it controlled the “doorway” that connected the worlds of China and those of India and the Arab world. Indeed, there are several places in Arabic texts where they state precisely that, namely, that on one side of Zabag/Zabej/Djawaga was the Chinese world and on the other side was the Indian/Arab world.
For many centuries, this was probably an easier way to get goods through the region than by going by sea down through the Straits of Melaka. However, to be able to move your goods across the Malay Peninsula, you had to be on good terms with Zabag/Zabej/Djawaga. This, I would argue, is why Arabic sources claim that Zabag/Zabej/Djawaga controlled many different “islands.”
Where were those islands? The geographic information in Chinese sources indicates that from Shepo, which I fully believe was the same place as Zabag/Zabej/Djawaga, one could reach the sea in four directions.
To the east was Champa and China, and those two names get mentioned frequently in Arabic sources in association with Zabag/Zabej/Djawaga. As for the other directions, I think that they pointed to various places on the western side of the Malay Peninsula and including the northern parts of Sumatra.
How far north up the coast were the areas that Zabag/Zabej/Djawaga controlled? I’m not sure. However, I think scholars should re-examine the Arabic sources and to do so realizing that Zabag/Zabej/Djawaga was in the area of Songkhla/Phatthalung/Sating Phra, not on the island of Sumatra.
Ok, so how did trading then work? While I think some people from India and the Arab world did trade by sailing down through the Straits of Melaka and then on to China, I think others journeyed to places like northern Sumatra and Kedah, and then either exchanged their goods there or had them delivered to Zabag/Zabej/Djawaga from where they were shipped off to China.
If that is the case, then who did this shipping? My answer: Chamic peoples.
There are a couple of issues that have long perplexed scholars. The first is the fact that the Acehnese language of northern Sumatra is a Chamic language. How did that happen? There are scholars who have pointed to the movement of small numbers of Cham refugees in the fifteenth and seventeenth centuries to places like Aceh, but that is insufficient for Acehnese to develop as a Chamic language.
The second is the fact that during the Song period the term “Pu” 蒲 can frequently be found in Chinese sources at the beginning of the names of envoys coming from certain places in Southeast Asia. There have been scholars who have tried to argue that this term indicates the Muslim name “Abu” and that these were perhaps Persians or Arabs.
Other scholars, meanwhile, have long noted that this could represent the Cham honorific term “po.” However, because such scholars have also believed that places like “Srivijaya,” which during the Song period has many envoys named “Pu/Po,” actually existed and was on the island of Sumatra, and that Zabag/Zabej/Djawaga did as well, they have only been able to state that there must have been close ties between Champa and (the imagined) “Srivijaya.”
What I see, however, is something very different. I think that Chamic peoples must have been involved in the trans-peninsular trade at Shepo/Zabag/Zabej/Djawaga. Over time they probably became active on both sides of the peninsula, and this would explain why Acehnese is a Chamic language.
Take for instance the following (and I can provide many similar examples): In 1015 the king of the Chola kingdom in India sent a mission to the Song court. It was led by the envoy Suolisanwen 娑里三文. The deputy envoy was Pu [Po] Shu 蒲恕.
If Chamic peoples were running much of the trans-peninsular trade at Shepo/Zabag/Zabej/Djawaga, it would make perfect sense that a ship from the Chola kingdom would sail to northern Sumatra or Kedah, and that after traveling overland to Shepo/Zabag/Zabej/Djawaga, the envoy would then board a Cham ship that would take him to China.
The inexperienced Chola envoy would therefore have relied on a Cham voyager who was more familiar with the journey to China, and that man would have become “deputy envoy.”
There is a lot more that I can say about this, but I think I’ve made my point. When we “re-arrange” ourselves and see Shepo/Zabag/Zabej/Djawaga as a place that controlled the trans-peninsular trade at Songkhla/Phatthalung/Sating Phra (rather than a place on Sumatra) then a lot of related historical details suddenly make sense.
There must be many more of those details in Arabic sources, so for whoever is interested, there is a lot of work to do with those historical materials!!
This Post Has 4 Comments
So that is why the chams early on could take on the vietnamese, they controlled a trade route and alot of wealth. as soon as they lost the trade route they quickly succumbed to the vietnamese.
the rise of ayutthaya caused the down fall of the khmer and chams as they disrupted the traditional trade routes. another reason i refuse to eat pad thai lol. just kidding.
I think “thalaj” is a Mon-Khmer word which means “river”, “a place where many rivers meet”, or “a large body of water”. In Laven language (Bahnaric): [təlaj] = river. In Kui language (Katuic): [tlee] = large river. In Nyah Kur language (Monic): [thəlee] = area where several rivers meet. Proto Mon-Khmer: [*d[n]liʔ] = large river, sea.
Thank you!! Yes, I looked this up too after writing that post, and I agree with you.