In order to help readers understand how I came up with the idea that there was no maritime empire called “Srivijaya” and that the main place name in Chinese sources that people think refers to “Srivijaya” – Sanfoqi – actually refers to “Kambuja” (Angkor), in this post I’m going to outline the steps I took to reach this conclusion.
Step One: Investigate Names
The first thing I did was to try to figure out what Shilifoshi and Sanfoqi might have meant. As I stated, I’ve always been unsatisfied with the claim that those two terms referred to a supposed maritime kingdom based at Palembang, Sumatra called “Srivijaya” because there is absolutely no possible way that those terms can mean “Srivijaya.”
So what do they mean? In the first post I came up with the idea that Shilifoshi could have meant something like “Sri Budhjeta,” and I talk about where it might have been.
Ultimately, however, the issue of what and where Shilofoshi was is of secondary importance, because the other term is the one that has been used the most to develop a history for “Srivijaya.”
I then started to look into the meaning of Sanfoqi. I checked what those three characters might have sounded like 1,000 years ago, and came up with something like “Samfhutshiaj.”
I then realized that this was similar to the term “Jianpuzhai” which is mentioned as a name for what we think of as “Cambodia” in the Mingshi (History of the Ming). This started to get me thinking about the possibility that Sanfoqi might be referring to something like Kambuja (See the second post).
Many scholars have talked about place names (although I don’t know of anyone who has interpreted the name Shilofoshi as I have nor have I ever seen anyone equate Sanfoqi with Kambuja as I do). However, what I did following my examination of names (keep reading below) is something that I have not seen any other scholar do. As such, I believe that I am putting forth a new argument.
Step Two: Test the Hypothesis that Sanfoqi = Kambuja
In the third post, I tested the hypothesis that Sanfoqi = Kambuja. There is a term in Chinese sources, Zhenla, that people think refers to “Cambodia,” I wondered if Zhenla might be a reference to the area of the Mekong Delta from the coast up to say Phnom Penh, or perhaps even the Champassak area of southern Laos and northern Cambodia, whereas Sanfoqi might refer to Angkor.
So I pictured four “kingdoms” (and yes, that term is in scare quotes because we are of course talking about fluid “Mandala polities”) next to each other: Siam, Sanfoqi, Zhenla, and Champa. I then checked the information in the dynastic histories and the Ming shilu (Veritable Records of the Ming), and in particular, I looked for information about contact between these kingdoms so that I could see if it made sense to think of Sanfoqi as Angkor, and as a polity in interaction with Siam and Zhenla.
We know, for instance, that Siam started to attack Angkor in the late 1300s, and we also know that at least one Angkorean king fled eastward. I looked at the information about Siam to see if there was anything about that. There was.
I then looked at the information about Sanfoqi to see if any of the information there corresponded. It did. Then I looked at information about Zhenla to see if I could detect an Angkorean king relocating there, and I could.
So from that test, I could clearly see that it made sense to think of Sanfoqi as Angkor.
Step Three: Test the Details
Having made that connection, I then looked through the information about Siam, Sanfoqi, Zhenla, and Champa to see if the detailed information there supported or contradicted this claim. I was particularly interested in details that had long confused scholars, like statements to the effect that the clothing of people from Champa was like that of Arabs, and that the people in Sanfoqi were all surnamed “Pu,” a term that some historians think was a Chinese rendering of the Arabic “Abu.”
Posts eight, nine, and eleven all examine such details, and they all support the idea that Sanfoqi = Angkor. Further, I also found that certain details that have long confounded scholars who have assumed that Sanfoqi = Srivijaya now make perfect sense. Finally, I did not find information that contradicts the idea that Sanfoqi = Angkor.
Step Four: Investigate Zhenla as “Cambodia”
As I stated above, the term “Zhenla” has long been seen by historians as referring to “Cambodia,” and there is a Chinese record of a journey to “Zhenla” that is clearly about a trip to Angkor.
So in the fourth post, I investigated this issue and came to conclude that Zhenla was a kind of “Chamic” polity in the Mekong Delta that served as the gateway to Angkor and where some members of the Angkorean ruling family (who traditionally referred to their realm as “Kambuja”) migrated to during the tumultuous years of the Siamese conquest of Angkor.
Step Five: Examine Why the “Srivijaya” Idea Emerged in the First Place
In the midst of this research, I also went back to look at the 1918 article by George Cœdès to examine his evidence for the idea that there was a maritime polity in Southeast Asia called “Srivijaya” and that the Chinese terms Shilofoshi and Sanfoqi signified that polity.
Posts five and six examine how Cœdès used a Sanskrit grammar argument to claim that there was a place in Southeast Asia called “Srivijaya.” In these posts, although my knowledge of Sanskrit is admittedly limited, I argue that there were serious problems with that claim and with certain aspects of Cœdès’s reading of Sanskrit texts, particularly when it came to identifying “kings” and “kingdoms.”
My reading of Cœdès demonstrated clearly to me that he never made a convincing argument for “Srivijaya.”
Finally, while I indicate in this post the topics that individual posts in this series cover. In reality, each topic is not restricted to an individual post. For instance, while each post focuses on an individual topic, there can be bits of information in other posts that support the arguments in other individual posts. As such, the overall argument here is best understood by reading all of the posts.
