Srivijaya 3.0 (05): Ok, So There WAS a Srivijaya. . . But It’s Not What You Think. . .

A while ago I made a video where I said something like “no place called Srivijaya ever existed”. . .

I kind of knew at the time that I might end up eating those words, but I like to be provocative, and I had a strong sense at that time that the presence of the term “Srivijaya” in some stone inscriptions could be interpreted as the title of a person.

Now I think differently, but to explain what I now think, requires quite a bit of explanation. So, here we go.

The Background

In 1918, French scholar George Cœdès wrote an article in which he claimed that there had been a kingdom in the past called “Srivijaya” that was located around the area of Palembang on the island of Sumatra, and that Chinese sources referred to it as Shilifoshi and Sanfoqi.

From that point to the present, scholars have been interpreting historical sources and archaeological information based on that claim.

A couple of years ago I realized that those two places in Chinese sources do not refer to a kingdom on the island of Sumatra called “Srivijaya.”

I originally wrote about Shilifoshi here:
https://leminhkhai.blog/was-there-a-srivijaya-im-not-sure-but-there-was-definitely-a-sri-budhjeta/

But I have updated my ideas, and now think that Shilifoshi definitely was in the area of what is now Southern Thailand.

I posted about that here and here.

And I have published an article on Sanfoqi here, where I demonstrate that this term referred to “Kambuja,” and in particular, the place that we today think of as Angkor:
https://www.academia.edu/83822426/Rescuing_History_from_Srivijaya_The_Fall_of_Angkor_in_the_Ming_Shilu_Part_1_

Methodology

One of the great failings of the extant scholarship on “Srivijaya” is that it was constructed like a house of cards.

Cœdès did not know Chinese. His understanding of Sanfoqi was based on the earlier work of a scholar by the name of Willem Pieter Groeneveldt who had misinterpreted Chinese sources to argue that “Sanfoqi” referred to a place on the island of Sumatra around Palembang.

In other words, a core concept that the “Srivijaya” idea is based on, that this kingdom appears in Chinese sources as “Sanfoqi,” is false. In the century since Cœdès made that claim, countless books and articles have employed information about “Sanfoqi” to write about “Srivijaya.”

Therefore, this scholarship is now like a house of cards. When we pull out the “Sanfoqi = Srivijaya” card that serves as part of the foundation of the “Srivijaya house,” that house largely collapses.

In my article, I try very hard to avoid engaging in this approach. I do not make a claim and then interpret information based on that claim. Instead, I lay out the information about Sanfoqi, and then point out the places where we can see that this information is clearly not talking about the island of Sumatra and the places where we can see that this information is clearly talking about Cambodia/Angkor.

I think what I present there is a much more solid assessment of the historical information about “Sanfoqi” than has ever been presented before, and it can therefore serve as a foundation for building actual historical knowledge, rather than a house of cards.

Inscriptions

Meanwhile, what I have not previously written about in detail are inscriptions that mention the term “Srivijaya.” As I see it, there are three different types or groupings of these inscriptions.

First, there are some inscriptions that have been found in the area of southern Sumatra that date from the 680s. These inscriptions all seem to relate to the actions of a single ruler, and that ruler appears to have been attempting to expand or solidify his authority over that area. It looks like this guy was going out and conquering places and then erecting inscriptions that basically were intended to say “Don’t mess with Srivijaya.”

The first scholar to have examined one of these inscriptions believed that “Srivijaya” here referred to a person. Since I knew that the Chinese sources do not support the idea that there was a kingdom on Sumatra called “Srivijaya,” that idea made sense to me, and I thought that the term “Srivijaya” in these inscriptions likely indicated a person. That is why I said in a video that “no place called Srivijaya ever existed.”

However, recently a colleague who is very knowledgeable about Sanskrit inscriptions in Southeast Asia told me that it is unlikely that the term “Srivijaya” in these inscriptions refers to a person because it follows another term, “kadatuan,” which can refer to anything from a “royal residence” to a “kingdom,” and when we look at a comparable term in later Javanese inscriptions, “kraton,” that term is not normally followed by the name of a ruler.

I accept that. However, that then leaves us with a question: What was “Srivijaya”? Was it the name of a “royal residence” or a “kingdom”? Or perhaps something in between?

To answer that question, we need to look at the other inscriptions that mention “Srivijaya.” One is known as the “Ligor inscription,” and it was found in the area of what is now Nakhon Si Thammarat in southern Thailand, and it dates from 775, that is, roughly a century later than the earlier Sumatran inscriptions.

This inscription mentions “Srivijaya” and it also mentions a king called Sri Maharaja who was from the Śailendra “dynasty” or line of kings (Śailendravamśa). There is evidence of a “Śailendra dynasty” on the island of Java starting roughly around this time, and this Ligor inscription has led some scholars to speculate that there had once been a “Śailendra empire.” Meanwhile, since this inscription mentions “Srivijaya,” there are others who have argued that this inscription is an indication that the area of southern Thailand was part of a larger “Srivijayan” empire or world.

Ultimately, I don’t think that the Ligor inscription on its own can help us determine what “Srivijaya” referred to. For that, we need to turn to the third type of inscription, and that is inscriptions from the Chola kingdom in southern India.

There are a few Chola inscriptions that mention “Srivishaya,” rather than “Srivijaya,” however scholars argue that these two terms were interchangeable. I find two of these inscriptions to be particularly helpful for determining what “Srivijaya/Srivishaya” referred to.

One of the inscriptions mentions that the ruler of Srivishaya became the ruler of Kedah. The other inscription details a Chola attack against the king of Kedah and indicates that “Srivishaya” was captured and that it contained various gates and other prominent structures as if it were a kind of “capital.”

From these two inscriptions, one can gain a sense that there was a close relationship between Srivishaya and Kedah and that Srivishaya held a more important position.

New Claims

There are ways to interpret this information, but doing so requires employing a different methodology from the one I have employed to date. As stated above, my article on Sanfoqi just lays out the information and does not make a claim and then interpret information based on that claim. However, to understand the inscriptions requires that we examine the information in those inscriptions based on claims which I think the information in my article helps make evident, and which I think I have sufficiently documented.

What are those claims?

1) There is no support in Chinese sources for the existence of a kingdom in southern Sumatra called “Srivijaya.” Further, when we realize that “Sanfoqi” refers to “Kambuja,” what we find is that there is very little information at all in Chinese sources about the area of Java, southern Sumatra, and the Straits of Melaka until the Mongol and Ming periods (1271-1644).

2) There is a place in Chinese sources, Shepo, which scholars have long believed is an earlier name for “Java.” I demonstrate in my article that this is not true. The confusion stems from the fact that historically these two terms sounded similar. A thousand years ago, the characters that are pronounced now in modern Mandarin as “Shepo,” were used to represent a name that sounded something like “Jaba.”

This place, Shepo/Jaba, disappears from Chinese sources in the early Ming period, and in late-Ming and early-Qing sources, the idea emerged that this name was an earlier name for Java. However, there is evidence, which I present in my article, that demonstrates that this is not true, and we can also see from geographical descriptions about Shepo/Jaba that it is not referring to the island of Java. Instead, I locate Shepo/Jaba in the area of what is now Songkhla in southern Thailand.

New Ideas Based on the Above Claims

The above geographical “re-arrangement” which the information in my article supports is what I am now using to understand and interpret other information. So, in what follows, I will make arguments based on the claims that Sanfoqi/Kambuja was “Angkor” and Shepo/Jaba was “Songkhla.” Having made a thorough examination of the information in Chinese sources about those places, I am confident that I am not constructing a house of cards, as Cœdès did when he claimed that “Srivijaya” was referred to in Chinese sources (which he could not read and had to rely on the misunderstanding of Groeneveldt) as “Sanfoqi.”

1) What becomes obvious when one understands that Sanfoqi/Kambuja was “Angkor” and Shepo/Jaba was “Songkhla” is that certainly during the Song Dynasty period (960-1279), but likely starting earlier, the main international trade route between the Middle East/India and China was across the Malay Peninsula. Arabic sources, for instance, make it clear that most traders from the Middle East journeyed to Kedah or northern Sumatra, and then stopped there.

Those sources also make it clear that Kedah and certain places in northern Sumatra were “vassals” of a place called “Zabag.” This place, “Zabag,” is described as a kingdom and an empire. It is also how the city where the ruler of this kingdom/empire lived is called, and that ruler is referred to as the Maharaja, the same term that appears in the Ligor inscription.

I think what Arabic sources reveal here is a classic description of a “mandala” polity, that is, of a place that has a powerful center that claims authority over “vassal” polities.

Based on geographical and other information in Arabic and Chinese sources, I believe that Zabag was the place that Chinese sources refer to as Shepo/Jaba (indeed, at times in Arabic sources it is also referred to as “Jaba”), and the place that I argue was somewhere around what is now Songkhla, not all that far from where the Ligor inscription was found.

Chinese sources provide the distance from Shepo/Jaba to the sea in all four directions. Those sources reveal that immediately to the east was the sea, as this place was located on the coast. In the other directions, the sea could be reached the fastest towards the south, where it could be reached in five days. Further, these sources also say that this is where “Middle Eastern/Arab” peoples were.

All of this makes sense if we see Shepo/Jaba located at Songkhla and controlling the trans-peninsular trade route that led to places like Kedah, where “Middle Eastern/Arab” peoples stopped and unloaded their ships.

How, did their goods then make it to China? They were transported to Shepo/Jaba/Songkhla and then shipped to Cambodia, Champa, and China. There is quite a bit of evidence for this in Chinese and Arabic sources, and I’ll be writing in more detail about that soon.

2) So, people from the Middle East and India sailed to northern Sumatra and Kedah. Their goods were then delivered overland to Shepo/Jaba/Zabag/Songkhla. Someone then loaded these goods onto ships and took them to Cambodia, Champa, and China. Who did that? Probably various peoples, but Cham(ic) peoples definitely played an important role.

In Song Dynasty sources, many tribute missions from different places in Southeast Asia are delivered by peoples whose name begins with the character “pu” 蒲. These “pu” people also delivered tribute for people from as far away as India.

A lot of ink has been spilled debating what this means, with some arguing that it was the simplification of the Muslim name “Abu,” while others have claimed that it was a “Southeast Asian” term like the honorific “po” in Cham or a “pu” that appears in the middle of a certain Javanese title.

Actually, I would argue that the Cham honorific “po” is the only logical candidate for the source of this term. When we then consider that A) Acehnese, a language from northern Sumatra, one of the places where traders from the Middle East and India stopped, is a Chamic language, and that B) the main trade route went across the Malay Peninsula. . . I think it becomes very clear that a thousand years ago, Cham(ic) peoples were very active across a very wide swath of Southeast Asia.

3) Meanwhile, and I think very importantly, from Cham inscriptions, we find that the term “vijaya” was used as a kind of place name. Indeed, one of the meanings of “vijaya,” as well as “vishaya,” in Sanskrit is a “district.”

I need to learn more about how this term was used in Cham inscriptions, but a Vietnamese colleague who researches about Champa said to me that it is a term that functioned like the terms “pura” and “nagara,” that is, it referred to some kind of “center of power” and perhaps a fortified settlement.

My Current Interpretation

Based on all of the above information, what I would now argue is that at the core of the place which foreigners referred to as Shepo/Jaba/Zabag, was a kind of royal district that the rulers there referred to as “Srivijaya,” meaning something like “the great/divine district.” This would have been in keeping with the way that the term, “vijaya,” is used in Cham inscriptions, and that would make sense given the close contact between Cham(ic) peoples and Shepo/Jaba/Zabag.

Further, it also makes sense that the local term for a place, “Srivijaya” in this case, would be different from the term that foreigners used to refer to that same place, as that is a phenomenon that we see in other places as well. “Champa” is a perfect example of that, as there were various terms that the Cham used to refer to their polities over time whereas a single term, “Zhancheng,” continued to be used in Chinese sources for centuries.

Further still, it would also make sense that some people in the Chola kingdom would know the name, “Srivijaya,” as their inscriptions show that they were in close contact with its rulers. Indeed, the Chola kingdom attacked this Srivijaya, as well as other places, like its vassal, Kedah. There is not enough space to explain this here, but it looks to me that in the late-tenth and early-eleventh centuries the rulers of Shepo/Jaba/Zabag tried to gain more direct control over the trade network by attacking Angkor and ruling over Kedah themselves. It also looks to me like someone from Angkor may have convinced the Chola ruler to attack Shepo/Jaba/Zabag (or maybe he came up with this idea himself), and in doing so his forces conquered not only places like Kedah, but also “the great/divine district” of “Srivijaya” in Shepo/Jaba/Zabag.

Finally, from this perspective, it would also be possible to interpret the Sumatran inscriptions as having been left by the “Srivijaya” in Shepo/Jaba/Zabag.

As I stated above, those inscriptions appear to have been created by someone who was expanding his authority.

If Shepo/Jaba/Zabag was capable of attacking Angkor, one of the most populous settlements in the world at that time, it is then certainly conceivable that a Shepo/Jaba/Zabag ruler in the 680s could have sent people down into the Straits of Melaka to beat up on the comparatively smaller polities there as well, perhaps in an effort to make sure that trade did not pass through that region, but instead, went overland and through “the great/divine district” in the area of what is now Songkhla.

In other words, rather than imagining (based on a false claim) a “Srivijaya” at Palembang that extended its power up the Malay Peninsula, as some scholars have done, I would argue that the extant historical evidence provides much more support for the idea that there was a “great/divine district” called “Srivijaya” on the Malay Peninsula at the heart of the Shepo/Jaba/Zabag mandala and that rulers of this powerful center historically sought to extend their influence not only to Angkor but down to the southern end of the island of Sumatra as well.

So, yes, there was a “Srivijaya,” but it wasn’t the same “Srivijaya” that people are familiar with. Further, to see this “Srivijaya,” we need to relearn and reconceptualize many aspects of our understanding of Southeast Asian history.

Postscript

I had a colleague who is knowledgeable about Champa push back on what I said about “vijaya” as a place name meaning something like “district.”

As I said, I’m not sure about that.

But in thinking further, perhaps “Srivijaya” is similar to a name that appeared later “Nakhon Sri Ayutthaya.”

Most (or all?) foreigners referred to that place using some variation of the name “Siam.” And I was thinking, if we did not have access to Siamese sources (like we don’t have access to sources from Shepo/Jaba/Zabag), prior to the arrival of Europeans (who I think recorded the name “Ayutthaya” early on even though they ended up commonly referring to the place as “Siam”), we would have no idea that the name “Ayutthaya” even existed, or at the very least, would have extremely few references to it.

This might be the case with Srivijaya. It could be that this was the indigenous name for Shepo/Jaba/Zabag.

Actually, this is not all that different from the argument I made above, it’s just that the origin and usage of the term “Srivijaya” might not be related to the meaning of “vijaya” as “district,” but might be  based on its (more common) meaning as “victory.”